I

DEPRESSION IN CHILDREN AND ADOLESCENTS

Bwce E. Compas

Depression represents a significant mental health concern for children and adolescents. Research ---with both community and clinical populations has established that depressive sy~nptoms disorders and occur with increasing prevalence over the course of childhood and adolescence (e.g., Fleming & Offord. 1990; Petersen et al.. 19931. Moreover. depressive prdblems are associated wi'th significant impairment in the lives of children and adolescents-including disruption in academic achievement, peer relations, family functioning, and sense of self-and are strong predictors of referral for mental health services (Hammen & Rudolph, 1996). T h e tendency for depression to co-occur with other ~ r o b l e m and disorders further unders scores the importance of careful consideration of depressive problems in children and youths. Patterns of comorbidity suggest that a thorough understanding of depression in young people may facilitate knowledge of other problems and disorders as well (Compas & Hammen, 1994). This chapter provides an overview of concepts and methods for behavioral assessment of depression in young people. First, the nature of depression in young people is described with respect to three levels of depressive phenomena- depressed mood, depressive syndromes, and depressive disorders. This includes a brief discussion of current conceptual models of depression during childhood and adolescence, with emphasis on the need for a broad, integrative perspective, and consideration of developmental and normative processes. Second, the methods for the assessment of depression are critically reviewed, including self-report questionnaires, behavior checltlists, and diagnostic interviews. Third, the measurement of related constructs is reviewed, to provide guidance for the
~

assessment of depression in a broader context. Finally, the implications ofthis material for clinical assessment of and intervention for depression in young people are outlined.

NATURE OF DEPRESSION IN YOUNG PEOPLE

Nick is 15 years old and lives with his mother at the time that she brings him to the psychology clinic (his father left his mother before Nick was born). Prior to this visit, Nick was arrested for shoplifting at a local store. The store owner, a long-time friend of Nick and his mother, chose not to press formal charges, but expressed his concerns about Nick's behavior to his mother. She reports that the shoplifting incident was merely the "straw that broke the camel's back"-she has been very concerned about Nick for the past year. She reports that they are constantly arguing and fighting, and that the slightest problem seems to send Nick into a fit of anger. H e is irritable and sullen much of the time, and his angry outbursts have escalated recently, including several incidents in which he has thrown things and punched holes in the walls or doors at home. Nick's mother also reports that h e seems to be unhappy and withdrawn, spending more and more time at home alone. Her work schedule requires her to leave the house early in the morning, and Nick is expected to get himself up and off to school after she leaves. Nick has not been following through o n this plan, however, and he is now absent from school more days than not. An initial interview with Nick corroborates much of what his mother has reported. Nick is

198

P A R T 111. E M O T I O N A L A N D S O C I A L D I S O R D E R S
Taxonomy and Assessment

sullen and aloof during the interview, and shows little emotion. He describes a typical day as beginning in this manner: He wakes up early in the morning, after having stayed up late the night before watching television, but he lies in bed until 9:00 or 10:OO A.M. He then spends much of the day at home alone playing video games or watching television. He reports that he is having trouble controlling his appetite; he eats junk food and snacks all day long, and has gained considerable weieht as a result. His mother returns home from u work late in the afternoon, and they typically get into an argument about his having missed another day of school. They eat dinner together silently while watching television, and then spend much of the evening arguing about his homework, his school attendance problems, and his refusal to go to bed before midnight. Over the course of the interview, Nick reports that he is very unhappy, has very few sources of pleasure in his life, and feels hopeless that things could improve for him. Nick was born with a curvature of his spine; as a result, he walks awkwardly and is limited in his physical abilities. He reports that he is selfconscious about his appearance, feels that he is disliked and teased by his peers, and hates himself. Nick thus exemplifies many of the features that we associate with depression in young people -sad or dysphoric affect, irritability, disruption of sleep and appetite, social withdrawal, low selfworth, and a sense of hopelessness or pessimism about the future. One of the major challenges facing researchers and clinicians concerned with depression during childhood and adolescence involves operationally defining and measuring this construct. Several factors have i m ~ e d e d d e v e l o ~ m e nof meththe t ods to measure and understand depression in young people (Poznanski & Mokros, 1994). T h e first was the theoretical position, derived from the psychoanalytic perspective, that depression is not possible in children because of inadequate development of the superego. The second impediment was the belief that depression can occur but that it is masked by other characteristics or problem behaviors. For example, depression was often assumed to be masked by the presentation of disruptive behaviors. Both of these beliefs have now been thoroughly disproven by research verifying that children and adolescents do indeed ex~erience and clinically present with depression ( ~ a r l s o n & Cantwell, 1980; Hammen & Compas, 1994). Controversy continues, however, regarding the ways that depression is conceptualized in young people.

Current research and clinical ~ r a c t i c e concerning depression in children and adolescents have been hindered by two factors. First, researchers and clinicians have drawn upon different definiand different taxonomic svstions of de~ression tems, including a focus on (1) depressed mood, (2) empirically derived syndromes that include depressive symptoms, and (3) a constellation of symptoms meeting diagnostic criteria for a categorical disorder. These three approaches to depressive phenomena during childhood and adolescence have all been included under the general label of "depression," and this has led to confusion and miscommunication. Second, manv different types of assessment and diagnostic toolihave been used in the measurement of childladolescent depression. These measures have varied in the breadth versus specificity of the symptoms that are assessed; in their sources of information (children1 adolescents, parents, teachers, clinicians); and in their psychometric quality. This heterogeneity in the conceptualization and measurement of depression during childhood and adolescence has resulted in aYfragmentation of research efforts and has impeded determination of the prevalence of depressive phenomena, understanding of the developmental course of depression, and identification of etiological factors. Therefore, clarifying the relations among the three approaches is the first step toward understanding the nature of depression in childhood and adolescence. Defining and understanding childladolescent " depressionYare dependent on the paradigms one uses for the assessment and taxonomy of psychopathology. Broadly defined, "assessment" is concerned with the identification of distinguishing features of individual cases, whereas "taxonomy" is concerned with the grouping of cases according to their distinguishing features (Achenbach, 1985,1993).Assessment and taxonomv are linked to each other, in that the grouping &cases in a taxonomic system should be based on clearly defined criteria and procedures for identifying the central features that distinguish amone cases. Simiu " larly, assessment procedures should reflect certain basic assumptions of the underlying system for classifying the phenomena of interest. Three a ~ ~ r o a c h tos the assessment and taxe onomy of Kildladolescent psychopathology have been reflected in the literature (Angold, 1988; Cantwell & Baker, 1991; Compas, Ey, & Grant, 1993; Kovacs, 1989). T h e first approach does not involve a full taxonomic or assessment paradigm;

Chapter 4 Depressron In Chtldren and Adolescents

7 99

it is concerned with depressed mood or affect, as represented by the work of Petersen (e.g., Petersen, Sariglani, & Kennedy, 1991), Kandel (e.g., Kandel & Davies, 1982), and others. T h e study of depressed mood during childhood and adolescence has emerged from developmental research in which depressive emotions are studied along with other features of biological, cognitive, and social development. T h e second approach is concerned with syndromes of behaviors and emotions that reflect depression; depressive syndromes are identified empirically through the reports ofchildred adolescents and other informants (e.g., parents, teachers). This strategy involves the use of multivariate statistical methods in the assessment and taxonomy of child and adolescent psychopathology, represented by the empirically based taxonomy of Achenbach (1985, 1993). T h e third approach is based o n assumptions of a disease or disorder model of psychopathology; it is currently reflected in the categorical diagnostic system of the Diagnostic a n d Statistical M a n u a l of Mental Disorders, fourth edition (DSM-IV; American Psychiatric Association, 1994) and the International Classification of Diseases and HealthRelated Problems, 10th revision (ICD-10; World Health Organization, 1992). In additlon to reflecting different paradigms of assessment and taxonomy, these three approaches
I

to depression are concerned with different levels of analysis. T h e "depressed mood" approach is concerned with depression as a symptom or emotional state; the term "depression" refers to the presence of sadness, unhappiness, or blue feelings for a n unspecified period of time. No assumptions are made regarding the presence or absence of other symptoms (e.g., poor appetite, insomnia). T h e "depressive syndrome" approach is concerned with depression as a constellation of behaviors and emotions. "Depression" refers to a set of emotions and behaviors that have been found statistically to occur together in an identifiable pattern at a rate that exceeds chance, without implying any particular model for the nature or causes of these associated synlptoms. Differences between individuals are viewed in terms of quantitative deviations in levels of symptoms. An empirically deprived syndrome of depressive symptoms is best represented in the research ofAchenbach and colleagues (e.g., Achenbach, 1993). Most pertinent here is the syndrome 1abeledAnxious/Depressed, composed ofsymptoms reflecting a mixture of anxiety and depression (see Table 4.1). T h e syndrome has been replicated in large samples in both the United States and T h e Netherlands (Achenbach, Conners, Quay, Verhulst, & Howell, 1989; D e Groot, Koot, &Verhulst, 1994).

TABLE 4.1. Symptoms of the Anxious/Depressed Syndrome, Based on Parent (Child Behavior Checklist) and Adolescent (Youth Self-Report) Reports

Parent reportn Complains of loneliness Cries a lot Fears s h e might do something bad Feels s h e has to be perfect Feels or complains that no one loves himher Feels others are out to get himher Feels worthless or inferior Nervous, highstrung, or tense Too fearful or anxious Feels too guilty Self-conscious or easily embarrassed Suspicious Unhappy, sad, or depressed Worrying

Adolescent report6
I feel lonely I cry a lot I am afraid I might think or do something bad I feel that I have to be perfect I feel that no one loves me I feel that others are out to get me I feel worthless or inferior I am nervous or tense I am too fearful or anxious I feel too guilty I am self-conscious or easily embarrassed I am suspicious I am unhappy, sad, or depressed I worry a lot I deliberately try to hurt or kill myself I think about killing myself

I
printed by permission.

"The items in this column are from Achenbach (1991b, p 45). Copyright 1991 by Thomas M. Achenbach. Re-

bThe items In this column are from Achenbach (1991d, p 37). Copyright 1991 by Thomas M. Achenbach. Reprinted by permission

200

P A R T 111. E M O T I O N A L A N D S O C I A L D I S O R D E R S

A more "pure" depressive syndrome did not emerge in the reports of parents, teachers, and adolescents. The "categorical diagnostic" approach views depression as a psychiatric disorder. This approach not only assumes that "depression" includes the presence of an identifiable syndrome ofassociated symptoms; it also assumes that these symptoms are associated with significant levels of current distress or disability and with increased risk for impairment in the individual's current functioning (American Psychiatric Association, 1994). Differences among individuals are considered in terms of quantitative and qualitative differences in the pattern, severity, and duration of symptoms. With only a few exceptions, childladolescent depression is diagnosed according to the same DSM-IV criteria as adult depression. Depressive disorders are classified under the broad category of mood disorders and, to a lesser extent, under adjustment disorders. Key exclusionary criteria are that a diagnosis of mood disorder is not to be made if the symptoms are caused by an established organic factor or occur in conjunction with psychotic disorders (e.g., Schizophrenia, Delusional Disorder, Schize phreniform Disorder). Within the mood disorders, depression is divided into two categories: bipolar disorders and depressive disorders. In distinguishing between bipolar and depressive disorders, bipolar disorders are defined by the presence of manic or hypomanic symptoms that may alternate with depression. As this section emphasizes depressive disorders without manic or hypomanic symptoms, the reader is referred to the DSM-IV for more information regarding bipolar disorders (see also Carlson, 1994). T o meet the criteria for a Major Depressive Episode (MDE), the child or adolescent must have experienced five (or more) of the specified symptoms for at least a 2-week period at a level that differs from prior functioning, and at least one of the symptoms must be either (1) depressed or irritable mood, or (2) anhedonia (see Table 4.2). Irritable mood may be observed in lieu of depressed mood in childrenladolescents and is believed to be more common in this age group than in adults. A diagnosis of Major Depressive Disorder (MDD) is made when a childladolescent has experienced one or more MDEs and no Manic, Hypomanic, or Mixed Episodes. Diagnosis ofMDD in children and adolescents is further differentiated by severity and nature of the symptoms; a range of specifiers may be applied (see Table 4.21. The crilteria for a diagnosis of Dysthymic Disorder (DD) in childhood and adolescence are that

for at least a period of 1 year (as compared to 2 years for adults), a young person must display depressed or irritable mood daily without more than 2 months symptom-free, along with additional symptoms as specified in Table 4.3. There must be no evidence of an MDE during the first year of DD. D D is further classified by age of onset; occurrence of the disorder prior to 2 1 years of age is designated as Early Onset. Although a diagnosis of MDD takes precedence over a diagnosis of DD, childrenladolescents may be diagnosed with D D for 1 year and subsequent M D D if the more severe depression follows. These cases of NIDD juxtaposed with the history of more chronic and less severe D D are often referred to as "double depression." An additional diagnostic category of Depressive Disorder Not Otherwise Specified may be used to classify children1 adolescents with symptoms of MDD, DD, Adjustment Disorder With Depressed Mood, or Adjustment Disorder With Mixed Anxiety and Depressed Mood. The DSM-IV cites examples such as recurrent mild depression that is not severe enough to meet the M D D criteria.
Developmental and Normative Considerations

Interpretation of data obtained from these three approaches to conceptualizing depression are best considered in light of normative developmental data on the base rates of symptoms and disorders. In analyses of a single item reflecting unhappy, sad, or depressed mood, Achenbach (1991b) found that parents reported 10-20% of nonreferred boys and 15-20% of nonreferred girls as experiencing this symptom at least somewhat or sometimes during the previous 6 months. Adolescents' selfreports (Achenbach, 1991d) indicated that 2035% of nonreferred boys and 25-40% of nonreferred girls reported feeling sad or depressed during the prior 6 months. Gender differences were small in both parental and adolescent reports of this symptom, with girls scoring slightly higher than boys. Petersen et al. (1991) examined selfreports of depressed mood in a longitudinal study of 3 35 individuals from early adolescence through young adulthood. Depressed mood increased during adolescence for girls, but remained relatively stable for boys throughout adolescence. These same authors also found that reports of significant episodes of depressed mood (lasting 2 weeks or longer) increased from early adolescence (i.e., episodes occurring between sixth and eighth grades) to late adolescence (i.e., episodes occur-

Chapter 4. Depression in Children and Adohscents

201

TABLE 4.2. DSM-IV Criteria for Major Depressive Episode (MDE)

A. Five (or more) of the following symptoms have been present during the same 2-week period and represent
a change from previous functioning; at least one of the symptoms is either (1) depressed mood or (2) loss of interest or pleasure. Note: Do not include symptoms that are clearly due to a general medical condition, or mood-incongruent delusions or hallucinations. (1) depressed mood most of the day, nearly every day, as indicated by subjective report (e.g., feels sad or empty) or observation made by others (e.g., appears tearful). Note: In children and adolescents, can be irritable mood. (2) markedly diminished interest or pleasure in all, or almost all, activities most of the day, nearly every day (as indicated by either subjective account or observation made by others). (3) significant weight loss when not dieting or weight gain (e.g., a change of more than 5% of body weight in a month), or decrease or increase in appetite nearly every day. Note: In children, consider failure to make expected weight gains. (4) insomnia or hypersomnia nearly every day (5) psychomotor agitation or retardation nearly every day (observable by others, not merely subjective feelings of restlessness or being slowed down) (6) fatigue or loss of energy nearly every day ( 7 ) feelings of worthlessness or excessive or inappropriate guilt (which may be delusional) nearly every day (not merely self-reproach or guilt about being sick) (8) diminished ability to think or concentrate, or indecisiveness, nearly every day (either by subjective account or as observed by others) (9) recurrent thoughts of death (not just fear of dying), recurrent suicidal ideation without a specific plan, or a suicide attempt or a specific plan for committing suicide B. The symptoms do not meet criteria for a Mixed Episode.

C. The symptoms cause clinically significant distress or impairment in social, occupational, or other important
areas of functioning. D. The symptoms are not due to the direct physiological effects of a substance (e.g., a drug of abuse, a medication) or a general medical condition (e.g., hypothyroidism).

E. The symptoms are not better accounted for by Bereavement, i.e., after the loss of a loved one, the
symptoms persist for longer than 2 months or are characterized by marked functional impairment, morbid preoccupation with worthlessness, suicidal ideation, psychotic symptoms, or psychomotor retardation.
Note. From American Psychiatric Association (1994, p. 327). Copyright 1994 by the American Psychiatric Association. Reprinted by permission. When the criteria for MDE are used to make a diagnosis of Major Depressive Disorder (MDD), these specifiers can be used to describe the current MDE: Mild, Moderate, Severe Without Psychotic Features, Severe With Psychotic Features, In Partial Remission, In Full Remission, Chronic, With Catatonic Features, With Melancholic Features, With Atypical Features. (The remaining specifier, With Postpartum Onset, would not apply to children.) If the MDD is recurrent, these additional specifiers can be used: With artd Without Interepisode Recovery, With Seasonal Pattern.

ring between 9th and 12th grades) for both boys and girls, with a more pronounced effect for girls. Girls reported more episodes of depressed mood than boys at all age levels, with this gender difference increasing from early to late adolescence. There is now substantial documentation that children and adolescents who meet DSM-III-R or DSM-IV criteria for MDD and D D can be reliably identified in community samples (e.g., Fleming & Offord, 1990). Ten studies of MDD in community samples of children and adolescents reported lifetime prevalence rates ranging from 0% to 31%,with a mean of 11%(Petersen etal., 1993). Prevalence is greater among high-risk groups, most notably children and adolescents whose parents

are depressed (e.g., Weissman et al., 1987).Point prevalence estimates have been provided by Lewinsohn and colleagues from a large community sample of adolescents (Lewinsohn, Rohde, Seeley, & Hops, 1991; Rohde, Lewinsohn, & Seeley, 1991). On the basis of structured diagnostic interviews, 2.9% of a sample of 1,710 adolescents received a current diagnosis of either MDD, DD, or comorbid MDD and D D (Lewinsohn et al., 1991). Lifetime prevalence of depressive disorders was 20% in this sample-a finding within the range oflifetime prevalence rates in the earlier studies reviewed by Fleming and Offord (1990). In further analyses with this sample, Lewinsohn, Hops, Roberts, Seeley, and Andrews (1993) found

202

P A R T 111. E M O T I O N A L A N D S O C I A L D I S O R D E R S TABLE 4.3. Criteria for Dysthymic Disorder (DD)

A. Depressed mood for most of the day, for more days than not, as indicated either by subjective account or
observation by others, for at least 2 years. Note: In children and adolescents, mood can be irritable and duration must be at least 1 year.

B. Presence, while depressed, of two (or more) of the following:

(1) poor appetite or overeating (2) insomnia or hypersomnia (3) low energy or fatigue (4) low self-esteem (5) poor concentration or difficulty making decisions (6) feelings of hopelessness C . During the 2-year period (1 year for children or adolescents) of the disturbance, the person has never been without the symptoms in Criteria A and B for more than 2 months at a time. D. No Major Depressive Episode (see [Table 4.21) has been present during the first 2 years of the disturbance (1 year for children and adolescents); i.e., the disturbance is not better accounted for by chronic Major Depressive Disorder, or Major Depressive Disorder, In Partial Remission. Note: There may have been a previous Major Depressive Episode provided there was a full remission (no significant signs or symptoms for 2 months) before the development of the Dysthymic Disorder. In addition, after the initial 2 years (1 year in children or adolescents) of Dysthymic Disorder, there may be superimposed episodes of Major Depressive Disorder, in which case both diagnoses may be given when the criteria are met for a Major Depressive Episode. E. There has never been a Manic Episode . . . , a Mixed Episode . . . , or a Hypomanic Episode . . . , and criteria have never been met for Cyclothymic Disorder.

F. The disturbance does not occur exclusively during the course of a chronic Psychotic Disorder, such as
Schizophrenia or Delusional Disorder.

G. The symptonls are not due to the direct physiological effects of a substance (e.g., a drug of abuse, a
medication) or a general medical condition (e.g., hypothyroidism). H. The symptoms cause clinically significant distress or impairment in social, occupational, or other important areas of functioning.

Specify if:
Early Onset: if onset is before age 21 years Late Onset: if onset is age 21 years or older

Specify (for most recent 2 years of Dysthymic Disorder): With Atypical Features . . .
Note. From American Psychiatric Association (1994, p. 349). Copyright 1994 by the American Psychiatric Association. Reprinted by permission.

significant sex differences in point prevalence at the initial data collection, with girls qualifying for diagnosis more often than boys. However, at follow-up 1 year later, sex differences were nonsignificant. Sex differences for lifetime prevalence rates were significant, with girls showing higher rates than boys. Age effects and the interaction of age and sex were nonsignificant, although older adolescents were more likely to be diagnosed with DD. Early to middle adolescence is widely believed to be the developmental period when significant increases occur in depression and when girls

begin to experience significantly more depression than boys (e.g., Angold & Rutter, 1992; NolenHoeksema & Girgus, 1994; Petersen et al., 1991). However, documentation of the emergence of gender differences in depression during adolescence is most strongly supported by a significant interaction of age and sex-that is, evidence that girls become more depressed relative to boys as they grow older. Two reviews of the research on adolescent depression have revealed that despite the increasing theoretical and empirical work devoted to explaining age and sex differences in depression during adolescence, these differences are not as

Chapter 4. Depression in Children and Adokscents

203

robust as is widely assumed (Petersen, Compas, & Brooks-Gunn, 1992; Leadbeater, Blatt, & QuinIan, 1995). Specifically, many studies of depression during adolescence have not examined age and sex differences. Among studies that have examined such differences, findings have been inconsistent with respect to the main effects of age and sex, and, most importantly, the interaction of age and sex. For example, we (Petersen et al., 1992) reviewed 30 studies of "depressed affect" published between 1975 and 1991. Of the eight studies that tested age effects, six reported no main effects for age. Of the 13 studies that tested sex effects, 11 reported higher rates for girls, while 2 found no sex differences. Leadbeater et al. (1995) reported that 14 out of 21 studies found more depressive symptoms among girls than boys; 6 studies found no sex differences; and the remaining study found higher rates for boys than girls in a working-class sample. Neither we (Petersen et al., 1992) nor Leadbeater et al. (1995) reported findings with respect to age x sex interactions. Four recent community studies provide some clarification of the relation between age and sex in depression during adolescence. In a longitudinal com~nunity-based study, Ge, Lorenz, Conger, Elder, and Simons (1994) found a significant age x sex interaction for cross-sectional analyses of depressed mood, with the rate for girls increasing relative to that for boys between the ages of 13 and 16, and decreasing thereafter. Consecutive reports from analyses of the Dunedin, New Zealand longitudinal sample also showed an emergence of gender differences during adolescence. Girls were found to be four times less likely than boys to have an M D E at age 13 (Kashani et al., 1987); however, when examined 2 years later at age 15, girls were 1.8 times more likely to have an M D E than boys (McGee et al., 1990). Similarly, using retrospective reports of 1&year-old high school students, Giaconia et al. (1993) found that M D E was most likely to occur between the ages of 14 and 17 (age effect), and that girls were at much greater risk for an M D E during this high-risk period than were boys (age x sex interaction). Most recently, we (Hinden et al., 1996) examined age and sex differences in a nationally representative sample of children and adolescents with respect to depressed mood versus the Anxious/Depressed syndrome, parent report versus youth self-report, and cross-sectional versus longitudinal research designs. Overall analyses revealed expected sex differences, with girls showing higher levels of depression than boys across operational definitions, informants, and designs. Age differences and the

interaction of age and sex were less consistent, however, and differences overall were of small magnitude. In a complementarystudy, we (Cornpas et al., in press) found that gender differences in depressed mood, the Anxious/Depressed syndrome, and an analogue of M D D were more consistent and significantly larger in magnitude in a sample of adolescents who had been referred for mental health services than in a nonreferred sample. Thesc findings suggest that the emergence of gender differences in depressive symptoms in adolescence may be limited to a subgroup of adolescent girls, who represent an extreme ofthe distribution of depressive symptoms among the adolescent population.

Conceptual Models of Depression in Young People

T h e nature, etiology, correlates, and developmental course of depression in children and adolescents have been viewed from a wide range of theoretical perspectives. These include psychodynamic, behavioral, cognitive, interpersonal, family, biological, and environmental models that vary in their comprehensiveness and in their level of empirical support. A review of these various models is beyond the scope of this chapter (see Hammen & R u d o l ~ h , 1996, for a discussion of these models). The'iAplicaion of these models for the assessment of depression in young people is clear: Each model has led to the development of measures of constructs that are hv~othesized to be related to depression (these measures are reviewed below). O n the other hand, these various conceptual models have not influenced the assessment of depression per se, as the assessment of depression is guided by the taxonomic approaches described above. It has been argued that a broad integrative approach will prove more fruitful than will viewing these various conceptual models as competing explanations of depression in children and adolescents. As Hammen and Rudolph (1996) state, "These models share many common features: the contributions of earlv familv socialization to subsequent functioning, the emergence of internal representations or working models of relationships, the interplay between individual vulnerabilities and external experience, and the role of depression as both a consequence of prior psychological disturbances and as a risk factor for future difficulties" (pp. 184-185). Integration of these various perspectives has led to a developmentalb i ~ p s ~ c h o s o c i perspective on depression dural
z L

204

P A R T 111. E M O T I O N A L A N D S O C I A L D I S O R D E R S

ing childhood and adolescence (e.g., Cicchetti, Rogosch, & Toth, 1994; Gotlib & Hammen, 1992; Hammen & Rudolph, 1996; Petersen et al., 1993). Integrative models have important implications for the assessment of depression in young people. First, they emphasize that developmental processes and a young person's developmental level must be taken into account in the assessment process. Second, they highlight the need to assess a range of factors that may be associated with depression in children and adolescents; these include internal characteristics of the child or adolescent, as well as features of his or her social context. Third, integrative models take into account that the interplay among these factors and their salience may change with development. These issues are considered in more detail below.

ASSESSMENT OF DEPRESSION
The assessment of depression in children and adolescents has drawn on three primary methodologies-self-report questionnaires, behavior checklists, and structured diagnostic interviews. Each of these methods is reviewed, along with its relationship to the taxonomic perspectives (mood, syndrome, disorder) described above. The integration of these methods is also considered.

Self-Report Questionnaires
All individuals experience periods of sadness, unhappiness, or dysphoric mood at various points in their lives. These periods of depressed mood may occur in response to a variety of environmental and internal stimuli, may last for varying lengths of time, and may be associated with either few or many emotional and behavioral correlates. One approach to research on depression during childhood and adolescence takes depressed mood and associated symptoms as its central focus. From this perspective, no attempt is made to catalogue the features of a depressive disorder, nor are any assumptions made about the underlying etiology of depressed mood. Because depressed affect is an internalizing problem that may not be readily observable to others, several different measures have been developed to obtain children's and adolescents' reports of their negative mood. Self-report measures have included single items, scales designed specifically to assess depressed mood, and subscales of existing measures. Numer-

ous measures of child/adolescent psychological distress and psychopathology contain single-item indices of depressed mood. For example, the measures developed by Achenbach and colleagues (see "Behavior Checklists," below) all contain an item for "unhappy, sad, or depressed affect. Numerous self-reportscales and subscales have been used to assess self-reports of symptoms of depression, including the Children's Depression Inventory (CDI; Kovacs, 1980); the Beck Depression Inventory (BDI; Beck, Ward, Mendelson, Mock, & Erbaugh, 1961); the Reynolds Child Depression Scale (RCDS; Reynolds, 1989) and the Reynolds Adolescent Depression Scale (RADS; Reynolds, 1986); the Children's Depression Scale (CDS; Tisher & Lang, 1983); the Center for Epidemiologic Studies Depression Scale (CES-D; Radloff, 1977); the Emotional Tone scale of the Self-Image Questionnaire for Young Adolescents (SIQYA; Petersen, Schulenberg, Abramowitz, Offer, & Jarcho, 1984); and the Kandel Depression Scale (KDS; Kandel & Davies, 1982). The BDI and CES-D were developed for adults and have been applied without adaptation to adolescents; the other scales were all designed specifically for use with children or adolescents. Scales used to measure depressed mood share several common features (see Reynolds, 1994, for a review of self-report scales of child/adolescent depression). They all include lists of emotions and symptoms that reflect the central features of depressive disorders. For most ofthe measures, these symptoms are at least loosely based on the Research Diagnostic Criteria (RDC) for depression (Spitzer, Endicott, & Robins, 1978a, 1978b), the DSM, or measures of adult depression (e.g., the BDI; Beck et al., 1961).An exception is the Emotional Tone scale, which is part of a measure of adolescent self-image (the SIQYA; Petersen et al., 1984).The measures use either a Likert-type scale for respondents to rate the degree to which each symptom applies to them, or choices from a series of responses for each item reflecting varying levels of the severity of that symptom. The CDI is noteworthy because of its wide usage in studies of clinically referred and nonreferred children and adolescents, and because the full scale includes a wide range of symptoms other than depressed mood (see Table 4.4 for sample items and format of the CDI). A comprehensive factor-analytic study (Weiss et al., 1991) identified a subset of items reflective of depressed affect through reanalysis of data collected from five previous studies on samples of clinically referred

Chapter 4. Depression in Children and Adoles~ents TABLE 4.4. Sample Items from the Children's
Depression Inventory (CDI)

205

Pick out the sentences that describe your feelings and ideas in the past two weeks.

1. -I am sad once in a while

2.
-I am sad many times -I am sad all the time -Nothing will ever work out for me -I am not sure if things will work out for me -Things will work out for me okay -I do most things okay -I do many things wrong -I do everything wrong -I have fun in many things -I have fun in some things -Nothing is fun at all -I am bad all the time -1 am bad many times -I am bad once in a while -I think about bad things happening to me once in a while -I worry that bad things will happen to me -I am sure that terrible things will happen to me -I hate myself -1 do not like myself -I like myself

3.
4.

5.
6.

7.

Note. From Kovacs (1980). Copyright 1980 by Scandinavian University Press. Reprinted by permission.

children (n = 515, ages 8-12 years) and adolescents (n = 768, ages 13-16 years). Five factors were identified in separate factor analyses of children's and adolescents' responses; the first of these was labeled Negative Affect With Somatic Concerns for both age groups (the remaining factors differed somewhat for the two groups).This factor included the items "I feel like crying every day," "I am sad all the time," and "Things bother me all the time" for adolescents, whereas for children only the sadness item loaded on this factor. Somatic items also loaded on this factor, including "I worry about aches and pains all the time," "Most days I do not feel like eating," "I have trouble sleeping at night," and "I am tired all the time." Although a Negative Affect With Somatic Concerns factor was identified in both children's and adolescents' responses, only items concerning frequent sadness, fatigue, and aches and pains were common to both age grodps. Frequent crying, being bothered by things, trouble sleeping, and poor appetite were unique to adolescents7 responses, whereas not having friends was the only item unique to children's reports. Thus, the responses ofchildren and

adolescents on the CDI contain a relatively distinct factor that reflects depressed mood along with somatic symptoms. The structure of the Negative Affect factor, however, appears to differ somewhat for children and adolescents. With regard to psychometric properties, all of the scales meet at least minimal criteria for internal-consistency reliability, test-retest reliability, and stability over moderate periods of time. For example, the EmotionalTone scale has shown adequate internal consistency for boys (alpha = .81) and girls (alpha = .85) (Petersen et al., 1984), and the KDS has a similar level of internal consistency (alpha = .79; Kandel & Davies, 1982). Internal-consistency data have not been reported on the Negative Affect With Somatic Concerns scale of the CDI; however, internal consistency of the overall scale has been found to be adequate (e.g., alpha = 20; Smucker, Craighead, Craighead, & Green, 1986). Testretest reliability also tends to be adequate for these measures (e.g., r = .79 over 5-6 months on the KDS; Kandel & Davies, 1982). However, some research with the C D I suggests that depressed mood may be much less stable in nonreferred samples than in referred samples. For example, Saylor, Finch, Spirito, and Bennett (1984) found lower stability in the responses of a nonclinical sample (r = .38) than in those of an "emotionally disturbed" sample (r = .87). Establishing the validity of self-report measures has been more challenging, as external criteria for validation have been difficult to identify. For example, the KDS has been shown to correlate significantly with depressive items from the Symptom Checklist 90 (r = .92; Kandel & Davies, 1984). Although this provides a minimal check of the criterion validity of the measure, the meaning of this correlation is difficult to interpret, in light of the high degree of overlap in the items on the two scales and the fact that both measures were completed by adolescents at a single administration. Correlations between self-reports of depressed mood/symptoms and measures of constructs that are hypothesized to be related to depression represents another step in the validation of the former scales. Important correlates of self-reports of depressed mood and symptoms have included low self-esteem (e.g., Petersen et al., I984), hopelessness (e.g., Kazdin, Rogers, & Colbus, 1986), negative cognitions and cognitive errors (e.g., Leitenberg, Yost, & Carroll-Wilson, 1986), low levels of perceived control (e.g., Weisz, Sweeney, Proffitt, & Carr, 1993), and stressful life events

206

P A R T 111. E M O T I O N A L A N D S O C I A L D I S O R D E R S

(e.g., Compas, Grant, & Ey, 1994). Correlations ofthese other variables with depressed mood have typically been moderate, most often ranging from .30 to .50. The discriminant validity of self-report scales has proven much more difficult to establish (Kendall, Cantwell, & Kazdin, 1989). Measures of depressed mood are highly correlated with measures of other negative emotions, most notably anxiety (Brady & Kendall, 1992). This is not surprising, in light of the tendency of depressed emotions to load strongly on a broader factor of negative affect that includes sad, dysphoric, anxious, and angry affect (Finch, Lipovsky, & Casat, 1989; King, Ollendick, & Gallone, 1991; Watson & Clark, 1992). It appears that the most appropriate conclusion that can be reached about scales of depressed mood during childhood and adolescence is that they are reliable and valid measures of one component of negative affect, along with additional, heterogeneous symptoms.
Behavior Checklists

A second approach to the assessment ofchild and adolescent psychopathology is concerned with depressive phenomena as they relate to a wider range of other childladolescent problem behaviors and emotions (e.g., social withdrawal, attentional problems, aggression). This approach does nota priori assume the presence of an underlying structure of psychological disorders in childhood and adolescence. Rather, it is assumed that the structure and pattern of symptoms and disorders is best understood from data obtained from the most relevant informants about childladolescent behavior. In this sense, the identification of depressive syndromes is based on a deductive method, moving from the general (data on large samples) to the specific (the defining character2stics of an individual case). The three primay sources of information on childladolescent behavior that have been used most frequently are parents, teachers, and childrenladolescents themselves. Researchers taking this approach are faced with the task of aggregating and organizing the responses of large samples of respondents concerning the frequency and intensity ofa range of child1 adolescent behaviors and emotions. It is assumed that ifa disorder involves a syndrome of behaviors and emotions, it will be reflected in the statistical associations (intercorrelations) among problems that are reported as occurring together in samples of individuals. Specifically, ifthere is a syndrome ofdepression in childhood and adolescence, it will

appear as a set of recognizable problems that cooccur in reports by parents, teachers, andlor childrenladolescents. Various broadly focused parent, teacher, and self-report measures have been used to obtain reports ofbehaviors and emotions that are pertinent to the construct of childladolescent depression. Parent report measures have included the Revised Behavior Problem Checklist (Quay & Peterson, 1983),the Conners Parent Rating Scale (Conners, 1973), and the Child Behavior Checklist (CBCL; Achenbach, 1991b). Teacher checklists include the Conners Teacher Rating Scale (Conners, 1973); the Louisville Behavior Checklist (Miller, 1984); and the Teacher Report Form (TRF), a variation of the CBCL (Achenbach, 1 9 9 1 ~ ) . Multivariate measures of a wide range of behavioral and emotional problems that obtain adolescents' self-reports have been more limited. The only extensive analyses of adolescent self-reports have involved the Youth Self-Report (YSR), which is another variation of the CBCL (Achenbach, 1991d), and the Minnesota Multiphasic Personality InventoIy (MMPI; e.g., Archer, Pancoast, & Klinefelter, 1989; Williams & Butcher, 1989a, 1989b). These instruments differ from scales of depressed mood, in that they are designed to assess a wide range of internalizing and externalizing problems in addition to depression. Self-report behavior checklists have not been used with preadolescent children, as it is assumed that limits in young children's cognitive abilities and reading skills preclude the use of such measures with this age group (Achenbach, 1991d).Thus, the use of self-report measures of depressive syndromes applies only to adolescence. Depressive syndromes have been derived from factor analyses or principal-component analyses of the responses of large samples of clinically referred children and adolescents. The most extensive empirically based multivariate approach to the classification of child and adolescent psychopathology is represented by the ongoing work of Achenbach and colleagues. This system serves as the basis for discussion here, because (1) it is the only approach to date that involves the empirical integration of data from parents, teachers, and adolescents; (2) it is the only attempt to generate an empirically based taxonomy ofadolescent (and child) psychopathology; and (3) one of the measures (the CBCL) has been examined in a larger study of the Achenbach, Conners, and Quay Behavior Checklist (ACQ), in which items from other multivariate measures were also included (Achenbach, Howell, Quay, & Conners, 1991).

Chapter J Dtpression m Chrldren and Adoksre~tts

207

The use of the CBCL (Achenbach, 199Ib), TRF (Achenbach, 1991c),andYSR (Achenbach, 1991d) to generate a taxonomy of child and adolescent psychopathology is now in its second iteration. In both phases of the development of this taxonomy, principal-component analyses of checklist responses by parents, teachers, and adolescents have been used to identify sets of behaviors and emotions that co-occur in the reports ofthese informants. T h e question most pertinent here is whether a syndrome of behaviors and emotions has been identified that reflects depression. In the first stage of work on this taxonomy, data were analyzed separately as a function ofthe age and sex of the target children or adolescents and as a function of the source of information (Achenbach & Edelbrock, 1981). However, the use of different syndromes for different age groups, for boys and girls, and for different measures proved to be cumbersome in both research and practice. In the second generation of research on this system, steps were taken to integrate data from multiple sources and to identify core syndromes and cross-informant syndromes across ages, sexes, and sources ofinformation (Achenbach, 1991a, 1993) In principal-component analyses of CBCL, TRF, and YSR responses with samples of clinically referred children and adolescents, eight syndromes were found to be common across ages, sexes, and informants (these were labeled "crossinformant constructs"): AnxiousDepressed, Withdrawn, Somatic Complaints, Social Problems, Thought Problems, Attention Problems, Delinquent Behavior, and Aggressive Behavior. Most pertinent to depression, a syndrome of mixed anxiety and depressive symptoms emerged in these analyses (see Table 4.1 for the symptoms constituting this syndrome). A similar syndrome was identified in parents' reports from the larger item pool used in the ACQ (Achenbach et al., 1991). Thus, the characteristics of this syndrome are not limited to the items included only on the CBCL, YSR, and TRF; that is, the syndrome is not an artifact of the use of these scales. Correlations among different informants have been low in previous research, typically in the range of .20-.30 (Achenbach, McConaughy, & Howell, 1987). More recent research examining parent, teacher, and adolescent reports on the Anxious/Depressed syndrome has shown somewhat stronger correspondence amongthese reports than has been found on earlier syndrome scores (Achenbach, 1991a). Based on correlational analyses with referred and nonreferred adolescents, the mean of the cross-informant correlations (mother

x father, parent x teacher, parent x adolescent, teacher x adolescent) was .42 for both boys and girls. These correlations ranged from a low of .22 between reports of boys on the YSR and teachers on the TRF, to highs of .66 between mothers and fathers of girls and .70 between mothers and fathers of boys on the CBCL. Girls' reports on the AnxiousDepressed syndrome were correlated with their parents' reports (r = .44) at a significantly higher level than were boys' reports with their parents' reports (T = .32). Thus, moderate agreement was found across different informants. Although the scores generated from the most recent versions of the profiles for the CBCL, YSR, and TRF are moderately related to one another, they remain relatively distinct measures of anxious and depressed symptoms in adolescents. It appears best to continue to consider these as distinct perspectives on depressive syndromes in childhood and adolescence. The CBCL, TRF, and YSR all meet the necessary standards for acceptable internal-consistency reliability and test-retest reliability, and all are based on adequate norms from a nationally representative sample of nonreferred children and adolescents (Achenbach, 1991b, 1991c, 1991d). Internal-consistency reliabilities (as reflected by the alpha statistic) for the AnxiousDepressed scale on the CBCL,YSR, and CBCL were all above 35. One-week test-retest reliabilities were also strong, ranging from .74 for boys on the YSR to .87 for girls on theYSR and parents' reports ofboys' symptoms on the CBCL. The validity of the CBCL, YSR, and TRF has been established by comparing scores of referred and nonreferred youths. Referred youths scored significantly higher than nonreferred adolescents on the AnxiousDepressed scale on all three scales. Referral status accounted for 20% of the variance in AnxiousDepressed scores for adolescent boys and 22% for adolescent girls on the CBCL. Rates were lower on the YSR, with referral status accounting for 8% and 13% of the variance in AnxiousDepressed scores for boys and girls, respectively. Clinical cutoffs have been established that discriminate between referred and nonreferred youths. For example, 34% of referred adolescents scored above the cutoff for the AnxiousDepressed scale, as compared with only 5% ofthe nonreferred sample above this cutoff. This translates into a relative odds ratio of 10:l ( ~ . e .the odds of being , above the cutoff were 10 times greater for referred than for nonreferred youths). Other researchers have used items from these checklists to develop depression scales that are

208

P A R T 111. E M O T I O N A L A N D S O C I A L D I S O R D E R S

distinct from the Anxious/Depressed syndrome (e.g., Clarke, Lewinsohn, Hops, & Seeley, 1992; Connor et a]., 1996; Gerhardt, Compas, Connor, Hinden, & Achenbach, 1996; Nurcombe et al., 1989; Seifer, Nurcombe, Scioli, & Grapentine, 1989). These scales have been derived both empirically (Nurcombe et al., 1989) and rationally (Gerhardt et a]., 1996). For example, Gerhardt et al. (1996) identified a set of symptoms on the C B C L and YSR corresponding to the DSM-IV criteria for MDE, with matching items identified for all items except the one pertaining to anhedonia. The percentages (2.3% based on adolescents' self-reports, 1% based on parents7reports) of children and adolescents in a nationally representative sample who exceeded a cutoff for M D E (Gerhardt et al., 1996) were roughly comparable to the rate reported for community studies in which a diagnostic interview was used to derive diagnoses (e.g., Lewinsohn et al., 1993). These findings suggest that checklists can be used to generate a representative analogue of DSM symptoms of NIDE, with the caveat that they do not reflect DSM criteria for duration and level of impairment. In summary, the multivariate scales developed by Achenbach and colleagues meet the necessary psychometric criteria for use in the assessment of a depression-related syndrome labeled Anxious/Depressed. Although the scores generated from the most recent versions of the profiles for the CBCL, YSR, and T R F are moderately related to one another, they remain relatively distinct measures of anxious and depressed symptoms in adolescents. It appears best to continue to consider these distinct perspectives on the Anxious/Depressed syndrome in childhood and adolescence.
Structured Diagnostic Interviews

T h e categorical diagnostic approach to assessment of child/adolescent depression is represented in the use of structured or semistructured diagnostic interviews based on various versions of the DSM. T h e DSM-IV is based more upon the clinical literature of the symptomatology of disorders than on the empirical literature of the covariation of symptoms and syndromes. As such, it reflects an inductive approach, in which clinical information on specific cases is used to develop diagnostic decision rules that can be generalized to the larger population. The most extensively researched and widely used ofthese interviews are the Schedule for Affective Disorders and Schizophrenia for School-Age Children (K-SADS; Puig-Antich &

Chambers, 1978), the Child Assessment Schedule (CAS; Hodges, Kline, Stern, Cytryn, & McKnew, 1982),the Diagnostic Interview Schedule for Children (DISC; Costello, Edelbrock, Dulcan, Kalas, & Klaric, 1984), the Interview Schedule for Children (Kovacs, 1985), and the Diagnostic Interview for Children and Adolescents (Herjanic, Herjanic, Brown, & Wheatt, 1975). T h e present discussion focuses on the K-SADS, CAS, and DISC as representative examples of different types of diagnostic interviews for depression. T h e K-SADS was developed for youths 6-17 years old and was based on the aduIt Schedule for Affective Disorders and Schizophrenia (SADS). Although the SADS and K-SADS used the RDC for depressive disorders as a foundation, DSM-IV criteria may also be applied. The K-SADS is designed to be administered by a trained clinician to determine the onset, duration, and severity of current and past episodes of mood and anxiety disorders, Conduct Disorder, and psychotic disorders. Two versions ofthe K-SADS are available: one to assess a present episode (K-SADS-P), and an epidemiological version (K-SADS-E) to assess prior history. (See Table 4.5 for sample items from the K-SADS-P.) T h e semistructured assessment format consists of separate interviews with the youth's parents and the youth for approximately 1 hour each. Once the presence of a symptom has been established, the interviewer may use further questions to determine the severity and chronicity of the symptom. Ifresponses to initial questions are negative, the interviewer skips to the next section and thereby reduces the interview time for asymptomatic youths. Upon completion ofthe interview, the interviewer uses his or her clinical judgment to determine a summary diagnosis from the two sources of data (Chambers et al., 1985). Twelve summary scales, including four Depression scales, are generated and may be translated into diagnoses by the interviewer. Specifically, M D D , minor depressive disorder (currently a criteria set provided for further study), anxiety disorders, and Conduct Disorder may be diagnosed. More recently, the KSADS has been used to generate diagnoses of D D and Adjustment Disorder with Depressed Mood. The decision to convert scales and individual items into a diagnostic category is based upon the clinician's judgment rather than a computer algorithm (Edelbrock & Costello, 1988b). In addition, when .the parent's and the child/adolescent's accounts differ, the interviewer may meet with both parties to attempt to resolve the differences or make a judgment about which information to use in making a diagnosis (Chambers et al, 1985).

Chapter 4 Depression In Children and Adolestents

TABLE 4.5. Sample Items from the Schedule for Affective Disorders and Schizophrenia for School-Age Children, Present Version (K-SADS-P)

209

Note-If a mood disorder is present, the examiner should inquire if manic or hypomanic per~ods have or have not occurred. If a manic or hypomanic episode is suspected, the examiner should proceed . . . to establish presence or absence and the exact time periods Items [here] refer only to a depressive episode, not to a manic episode. Depressed Mood

3 Mild, e.g , often experiences dysphoric mood. It is not completely relieved by the presence of a
How have you been feel~ng? Have you felt sad, blue, moody, down, empty, like crying? (ASK EACH ONE) Is this a good feeling or a bad feeling? Have you had any other bad feelings? Have you cried or been tearful? Do you feel ( ) all the time, most of the time, some of the time . . . ? Does it come and go? How often? Every day? How long does it last? All day? How bad is the feeling? Can you stand it? What do you think brings it on? Do you feel sad when mother is away?

4 Moderate, e.g., most of the time feels

5 Severe, e.g , most of the time feels "wretched." 6 Extreme, e.g., most of the tlme feels extreme
depression which "I can't stand."

7 Very extreme, e.g , constant unrelieved, extremely painful feelings of depression.

SKIP T O BROODING, WORRYING

IF SEPARATION FROM MOTHER IS GIVEN AS A CAUSE: ) when mother is wlth you? Do you feel ( Do you feel a little better or is the feeling totally gone? Can other people tell when you are sad? How can they tell? Do you look different?
LASTWEEK: 0 1 2 3 4 5 6 7 WHAT ABOUT DURING T H E LAST WEEK? If the child has frequent dysphoric mood which is completely and totally relieved by the presence of the parent(s) (surrogate), check here Note-Sometimes the child will in~tially give a negative answer at the start of the interview but will become obviously sad as the interview goes on Then these questions should be repeated, eliciting the present mood and using it as an example to determine its frequency. Similarly, if the mother's report is that the child is sad most of the time and the child denies it, helshe should be confronted with the mother's opinion and helshe should be asked why does helshe think hislher mother believes helshe feels sad so often.
Note. From Pulg-Antich and Chambers (1978, p. 7)

The reliability of the K-SADS has been demonstrated more easily than the validity. Over a 72hour period, test-retest reliability of the K-SADS (as reflected by the kappa statistic) with 52 psychiatrically referred 6- to 17-year-oldswas established as on average .55 for individual items, .68 for sum-

mary scales, and .24-.70 for diagnoses (Chambers et al., 1985). Specifically, the four depression summary scales (Depressed Mood and Ideation, Endogenous Features, Depression-Associated Features, and Suicidal Ideation and Behavior) showed test-retest reliability above .67 and internal-

220

P A R T 111. E M O T I O N A L A N D S O C I A L D I S O R D E R S

consistency reliability above .68 (Chambers et al., 1985). Clinical syndromes and diagnoses are based upon the summary scores for articular symptoms found to be present. Agreement between parents and childrenladolescents was on average .53 for symptoms (range = -.08 to .96). T h e validity ofthe K-SADS is based on its ability to detect the prespecified diagnostic criteria and treatment effects (Edelbrock & Costello, 1988b). T h e K-SADS, for example, was sensitive to pharmacological treatment of affective disorders in preadolescent children (Puig-Antich, Perel, et a]., 1979). In addition, Puig-Antich, Chambers, et al. (1979) found that diagnoses from the R D C correlated with four case studies of biological correlates of depression. Finally, the K-SADS-P version was found to be correlated with the K-SADS-E administered 6 months to 2 years later about lifetime symptoms. Out of 17 children between the ages of 6 and 17, I 6 received the same diagnoses with both versions (Orvaschel, PuigAntich, Chambers, Tabrizi, &Johnson, 1982). No information is yet available on the instrument's discriminant validity, although Edelbrock and Costello (1988b) have suggested that the K-SADS may be able to differentiate between disorders through questioning about anxiety disorders, Conduct Disorder, and psychotic disorders in addition to depressive disorders. Finally, no epidemiological data have been povided for the purpose of normative comparisons.

Unlike the K-SADS, the CAS was developed from a traditional child clinical interview (Hodges &Cools, 1990). It is thematically organized around 11 topic areas: school; friends; activities; family; fears; worries and anxieties; self-image; mood (especially sadness); physical complaints; anger; and reality testing. (See Table 4.6 for sample CAS items.) Responses are scored on symptom scales and can be used to generate DSM-IV diagnoses according to detailed scoring instructions. Originally developed for use with children, the CAS has been successfully extended to adolescents (e.g., Kashani, Reid, & Rosenberg, 1989). Most pertinent to the present discussion are the Mood Content scale and the diagnostic scales for M D D and DD. T h e reliability of the CAS has been well established. Test-retest reliabilities of psychiatric inpatients over 9 days for the M D D scale (r = .89) and the D D scale (r = 3 6 ) were adequate (Hodges, Cools, & McKnew, 1989). Internal-consistency reliabilities of childladolescent and parent scores on the M D D and D D scales were also adequate. Internal consistency was higher for a psychiatric sample (.83 for M D D and .86 for D D for child1 adolescent interviews; .85 for M D D and .88 for D D for parent interviews) than for a community sample (.68 for M D D and .75 for childladolescent interviews; .55 for M D D and .67 for D D for parent interviews) (Hodges, Saunders, Kashani, Hamlett, & Thompson, 1990).

TABLE 4.6. Sample Items from the Child Assessment Schedule (CAS)

H. Mood and Behavior

Most children have a number of different moods or feelings. What kind of mood are you usually in? What is the worst you have felt lately?

( I f desired, ask about each response item as follows:) Would you say that you have been feeling: Mark "yes" if it applies: 142. sad? (or gloomy, blue, or down in the dumps)? 143. crabby, irritable, things get on your nerves easily? 144. feeling "empty" inside? 145. feel like you don't care about things anymore? 146. as if nothing is fun anymore? (If child responds "yes" to any of the feelings in #142-146 above, or indicates nontransient sadness, ask:) I have been asking you about your feelings. How much (often) have you been feeling this way? Feels sad nearly every day (or irritable, empty, hopeless, or has loss of pleasure). When you feel this way, do you feel this way while doing most everything you do, or just some things? Experiences sadness (or equivalents mentioned above) when doing all or almost all usual activities.
Note. From Hodges, Kline, Stern, Cytryn, and McKnew (1982, p. 186 ). Copyright 1982 by Plenum Publishing Corporation. Reprinted by permission.

Chapter 4 Depress~onrn Ch~ldren Adokscents and

211

T h e validity of the CAS has been examined through the comparison of childladolescent and parent responses, the association of the CAS w~th other clinical interviews, and the association ofthe CAS w ~ t h self-report questionnaires. Parent and child responses were moderately correlated on the M D D scale (r = .46) and the D D scale (r = .47) (Hodges, Gordon, & Lennon, 1990). Children and adolescents who received a diagnosis of M D D on the basis of the CAS differed from nondepressed youths on the C D I (Hodges, 1990) and, more specifically, on subscales of the CDI related to dysphoric mood, loss of personal and social interest, and self-deprecation (Hodges & Craighead, 1990). In contrast to the K-SADS and the CAS, the DISC may be administered by clinicians or lay interviewers because of its structured format. The DISC was developed as a youth version of the Diagnostic Interview Schedule for adults. Several revisions (DISC-R, DISC 2, DISC 2.1, DISC 2.25, DISC 2.3, DISC 3) have been reported in the literature (e.g., Bidaut-Russell et al., 1995; Fisher et al., 1993; Jensen et al., 1995; Shaffer etal., 1988,1992,1993; Schwab-Stone etal., 1993), and the DISC is likely to continue to undergo revision. Intended as a research tool for use in epidemiological studies of psychopathology, the DISC is a comprehensive approach to detecting the presence, severity, onset, and duration of a broad range of symptoms in 6- to 18-year-olds.The results may be presented in two ways: (1) as the number and severity of symptoms or (2) as scored DSM-IV diagnoses. DSM-IV diagnoses are developed from a set of operational rules on the level of symptoms needed to meet stringent criteria. T h e criteria are considered stringent, because

more than the mlnimum number of symptoms required to meet DSM-IV criteria must be present in order for the child to be assigned a diagnosis from the DISC (Costello, Edelbrock, & Costello, 1985). Diagnoses are generated from computer algorithms ofthe DISC items, and computer profiles are usually interpreted by trained clinicians. Hence the instrument has the capability to present symptoms and clinical disorders and syndromes based upon a set ofstringent diagnostic rules. T h e DISC consists of separate interviews with a parent and a childladolescent for approximately 60-70 minutes and 40-60 minutes, respectively. (See Table 4.7 for sample items from the DISC 2.3 for a child informant.) Reliability of the DISC has been shown through interrater reliability of lay interviewers (.98 o n symptom scores) and 2-week period test-retest reliability of the symptom scores (.76 on average for parents) (Edelbrock, Costello, Dulcan, Kalas, & Conover, 1985). Asample of 316 psychiatrically referred youths varied in terms ofthe reliability of their symptom scores over the 2-week period. O n average, adolescents (14-18 years old) were found to have higher test-retest reliability scores (.71) than younger children (.43). Interestingly, parental reports on adolescents were less reliable than those of preadolescent children. More recent reports suggest increased reliability with revised versions of the DISC (Schwab-Stone et al., 1993; Shaffer et al., 1992, 1993). Symptom scores from parent interviews detected DSM-I11 criteria moderately reliably (average kappa = .56, range = .35-.8 1). Less reliable diagnoses were derived from adolescent interviews (average kappa = .36, range = .12-.71). Interinformant agreement was higher for the 14-

TABLE 4.7. Sample Questions from the Diagnostic Interview Schedule for Children (DISC), Child Informant, Version 2.3

MDDIDD Now I'm golng to change the subject a bit and ask about some other feellngs kids sometimes have I'm going to start off with asking you about depression and feeling sad. 1. In the past 6 months, were there t~mes when you were very sad? I F YES A. When you feel sad thls way, does it last most of the day? B. Would you say you have been very sad a lot of the tlme for as long as a year? I F YES, C. Would you say most of the time? I F YES, D. Were you very sad most of the t ~ m e as long as 2 years? for E. Now thinking about just the last 6 months . . . Was there a time when you were sad almost every day? IF YES, F. Did this go on for 2 weeks or more?
Note From Shaffer et al (1992)

212

P A R T 111. E M O T I O N A L A N D S O C I A L D I S O R D E R S

to l&year-old group (r's = .29-.35) than for children, and higher for externalizing (behaviorlconduct) problems than for internalizing problems. The average correlations between separate parent and youth interviews was .27 (Edelbrock et al., 1985). Parent and child data are scored separately to yield diagnoses. Moderate diagnostic agreement has been reported for revisions of the DISC (Piacentini eta]., 1993). T h e question of the DISC's validity has been addressed by examining the instrument's ability to discriminate between youths referred for pediatric and psychiatric problems, as well as the DISC's correlation with parent and teacher ratings ofTotal Behavior Problems on the CBCL and TRF, respectively. Using the parent version of the DISC, Costello et al. (1985) found that 40 children (aged 7-1 1 years) referred for psychiatric problems scored significantly higher on all the symptom areas and the total symptom score than did 40 same-age children referred for pediatric problems. Significant correlations with the CBCL and T R F provide some support for the concurrent validity of the DISC (Costello et al., 1984). For example, the Total Behavior Problems score of the parent CBCL correlated .70 with the total symptom score of the DISC-R. Considering the evidence that there is often substantial parent-child disagreement on measures (Achenbach et al., 1987), it is not surprising that the child version of the DISC was correlated only .30 with the CBCL. Nonetheless, the internalizing disorders identified by the parent report on the DISC did not correlate more strongly with the TRF and CBCL Internalizing scale than with the Externalizing scale (Costello et al., 1985), suggesting that the DISC may not be as able to detect the internalizing (e.g., depressive) disorders. An alternative explanation is that there is a strong relationship between internalizing and externalizing disorders. Costello et al. (1985) found that the specificity of the DISC's parent version decreased from 95% to 30% when more than one disorder was present in their sample of psychiatrically referred children (40 children aged 7-1 1). Overall, structured interviews have several benefits. The foremost strength of structured interviews is that they constitute the only accepted method for deriving a DSM diagnosis of MDD or DD. In addition, Hodges (1990) and others argue that self-report questionnaires are often unable to differentiate between different disorders within a clinical sample. For example, the CDI failed to differentiate between adolescents with and without depressive disorders in a psychiatric sample

(Lipovsky, Finch, & Belter, 1989). Of course, the diagnoses of the adolescents being compared to the CDI results are drawn from diagnostic interviews. Finally, diagnostic interviews have shown sensitivity to change brought about by treatment. Nonetheless, there are some issues of concern with all the diagnostic interviews. As described by Hodges and Cools (1990), the use ofclinical judgment to make diagnoses when a parent and a youth provide discrepant information can be problematic. They cited the description by Poznanski, Mokros, Grossman, and Freeman (1985) of how diagnostic decisions in clinical research meetings represent more of a "battle of the wills" than a "meeting ofthe minds." Established guidelines for making these decisions are needed to improve the validity of these instruments, and such guidelines are currently being prepared for the DISC and other interviews (e.g., Piacentini et al., 1993).
Integration of Assessment Methods for Depressed Mood, Syndromes, and Disorders

The research and assessment methods reviewed above present a fragmentedpicture ofresearch and clinical practice pertaining to depression in young people. Given this rather disorganized state of affairs, can research and practice concerning depressed mood, depressive syndromes, and diagnostic categories of M D D and D D be integrated? These three approaches appear to be at odds with one another, as depression is conceptualized and measured differently within each approach. The differences between self-report measures and diagnostic interviews for adult depression have been examined in detail. One perspective is represented by Coyne and colleagues, who argue that selfreport measures of depressive symptoms reflect generalized psychological distress and not clinical depression (e.g., Coyne & Downey, 1991; Fechner-Bates, Coyne, & Schwenk, 1994). Selfreport measures for adults have only moderate to poor correspondence with diagnostic interviews for depression, as reflected in moderate rates of specificity (true positives) but poor sensitivity (true negatives). For example, in a study of425 primary medical care patients who completed the CES-D and a structured interview for the DSM-111-R, most subjects who scored high on the CES-D did not meet diagnostic criteria for MDD; a fifth of the patients with MDD had low scores on the CES-D; and the CES-D performed as well in detecting anxiety as in detecting depression (Fechner-Bates et al., 1994). Furthermore, the psychosocial fac-

Chapter 4. Depression in Children and Adokscentx

2 13

tors that correlate with self-report measures differ from the correlates of M D D (Coyne & Downey, 1991). Whether these same patterns are evident with children and adolescents is addressed below.
Comparison of Features of Depressed Mood, Syndromes, and Disorders

As a first step in comparing the core features of these three approaches to childladolescent depression, abbreviated versions of the items from a self-report questionnaire (CDI); items from the AnxiouslDepressed syndrome derived from the CBCL, YSR, and T R F (Achenbach, 1991a); and the symptom criteria for M D E from DSM-1V are presented in Table 4.8. Asterisked items (") are those that are common to at least two of the three approaches. Comparison of these items indicates only moderate overlap in the central features of these approaches. Only three symptoms-sad or depressed mood, feelings of worthlessness, and feelings of guilt-are common to all three approaches. Five items are common to the C D I and the Anxious/Depressed syndrome but are not included in the DSM criteria; six items are common to the C D I and the DSM but are not part of the Anxious/Depressed syndrome; and no items are included in both the syndrome and the DSM that do not also appear on the CDI. Finally, 13 items appear only on the CDI; six items in the Anxious1 Depressed syndrome have no corresponding items on the C D I or the DSM; and one item appears on the D S M criteria but not on the C D I or in the syndrome. T h e C D I offers good coverage of the DSM-IV criteria for MDE, with the exception of an item reflecting psychomotor agitation or retardation. In addition, the CDI oversamples some of the DSM criteria and includes several items that are not part of the criteria for M D E . For example, feelings of worthlessness are oversampled, as they are represented by five items. Items assessing noncompliance and fighting with peers are included on the C D I but are not part of the D S M criteria. T h e Anxious/Depressed syndrome differs from both the C D I and the DSM criteria for M D E by its inclusion of anxiety symptoms and by the exclusion of items representing the neurovegetative symptoms of depression (e.g., sleep disturbance, appetite problems, fatigue). T h e absence of these items is not the result of their absence from the measures used to derive the Anxious/Depressed syndrome, however, as the CBCL, YSR, and T R F include items representing all of these symptoms. It is noteworthy that these symptoms did not load

with the other core DSM symptoms (sadness, worthlessness, guilt) in the principal-component analyses used to derive the syndrome (Achenbach, 1991a). O n the other hand, symptoms characteristic of both depression and anxiety did form a reliable syndrome in the reports of parents, teachers, and youths. In this way, the Anxious/Depressed syndrome is similar to the broad construct of "negative affect" or "mixed anxiety-depression" (Watson & Clark, 1984,1992; Watson & Kendall, 1989). From the limited overlap among depressive symptoms in these various approaches, one would expect only low to moderate correspondence in the identification of de~ressive~ r o b l e m s when these approaches are applied to individual cases. Empirical studies of the correspondence among some of these approaches are now reviewed.
Empirical Studies of Sev-Report Questionnaires, Behavior Checklists, and Diagnostic Interviews

Correspondence among the three approaches to measuring child/adolescent depression has been examined by using diagnostic interviews to assign DSM-111-R or DSM-IV diagnoses and collecting questionnaire or behavior checklist data on the same sample of individuals. This method has been applied in four studies examining self-report inventories of depressive symptoms and diagnostic interviews (Garrison, Addy, Jackson, McKeown, & Waller, 1991; Gotlib, Lewinsohn, & Seeley, 1995; Kazdin, Esveldt-Dawson, Unis, & Rancurello, 1983; Roberts, Lewinsohn, & Seeley, 1991), and in three studies examining behavior checklists and diagnostic interviews (Edelbrock & Costello, 1988a; Rey & Morris-Yates, 1991, 1992; Weinstein, Noam, Grimes, Stone, & SchwabStone, 1990). Each study provides some evidence for the convergence of these different approaches to assessment of child/adolescent depression. Both Roberts et al. (1991) and Garrison et al. (1991) examined the relation between self-report inventories of depressive symptoms and diagnoses of depressive disorders based on clinical interviews in large community samples of children or adolescents. Garrison et al. examined the CES-D and the K-SADS; Roberts et al. examined both the CES-D and the BDI, along with the K-SADS. Both studies found that these scales served as useful screening instruments for diagnoses of M D D and D D , but that both the BDI and the CES-D produced substantial rates of false positives. T h a t is, most adolescents who received a diagnosis of

214

P A R T 111. E M O T I O N A L A N D S O C I A L D I S O R D E R S

TABLE 4.8. Comparison of Abbreviated Items from CDI; Core Anxious/Depressed Syndrome on CBCL, YSR, and TRF; and DSM-IV Criteria for MDE

CBCL, YSR, and TRF Anxious/Depressed 'Frequent sadness "Feels self 1s not as good as most 'Unhappy, sad, depressed *Feels worthless 'Feels too guilty "Feels lonely 'Bothered by things 'Frequent crying 'Worries bad things are going to *Lonely *Hurt when crlticlzed (TRF only) Cries a lot Worries 'Feels unloved
-

DSM-IV MDE *Depressed or irr~table mood 'Feel~ngs of worthlessness [or guilt] *Feelings of [worthlessness or] guilt
A

"Poor appetite "Trouble sleeping "Thinks about suicide

Doesn't have fun at school Doesn't have friends Feels inferlor to others Noncompliant Frequently gets into fights Hopelessness Incompetent Sees self as "bad" Self-hatred Doesn't want to be with people
-

'Diminished interest or pleasure 'Weight loss or gain "Insomnia or hypersomnia "Recurrent thoughts about death, suicidal ideation "Fatigue or loss of energy "Diminished abillty to think or concentrate
-

Fears own impulses Needs to be perfect Feels persecuted Fearful, anxious Self-conscious Suspicious
-

Psychomotor retardation or agitation

Note. Asterisked Items (*) are Included In two or more measures Dashes (-) Indicate that a corresponding Item was not found.

MDD or DD also reported elevated scores on the CES-D or BDI, but large numbers of individuals with elevated scores on the symptom measures were not judged to be clinically depressed on the basis of interview data. In a study of inpatient children, Kazdin et al. (1983) examined the corres~ondenceof child and parent reports on the CDI with interviews to assess depression. Withininformant responses on the CDI and the interview

were moderately correlated (r's = .62 for children, .71 for mothers, and .54 for fathers). Correspondence of the CDI and the interview across informants was low, however, with mother-child and father-child correlations (is = .16 and .33, respectively) lower than mother-father correspondence (r = .65). A recent study by Gotlib et al. (1995) provides the most comprehensive information on the as-

Chapter 4 Depress~ontn Chrldren and Adolescents

2 I5

sociation of self-report questionnaires and diagnostic interviews. In addition to examining the sensitivity and specificity of the CES-D and the K-SADS in a large community sample of adolescents, they carefully examined the characteristics ofthe false positives within the sample (i.e., those with high CES-D scores who did not meet diagnostic criteria for MDD). Adolescents classified as false positives manifested higher levels of current and future psychopathology than did the true negatives (those who scored low o n both the CES-D and did not meet criteria for NIDD). More importantly, the false-positive individuals did not differ significantly from the true-positive participants on a wide range ofmeasures ofpsychosocial dysfunction. T h e authors concluded that individuals who score high on a self-report measure of depressive symptoms but do not meet diagnostic criteria for M D D are far from a "normal" group; rather, these individuals are experiencing significant distress and impairment in psychosocial h n c tioning (Gotlib et al., 1995). T h e false positives differed from the true negatives on the number of DSM diagnoses they received, suggesting that the CES-D is a measure ofgeneral distress rather than depression specifically. However, the only specific diagnoses other than depression that distinguished the two groups were anxiety disorders. This elevation on diagnoses of depression and anxiety is consistent with the high rates of comorbidity of these two groups of disorders that have been reported in previous work (Compas & Hammen, 1994). Furthermore, those who met criteria for M D D also demonstrated a high rate of comorbidity of depression and anxiety. It appears, therefore, that the CES-D may be related more strongly to symptoms of depression and anxiety than to symptoms associated with other forms of psychopathology. Finally, the false-positive participants were at least twice as likely as the true-negative participants to develop a psychiatric disorder over the course ofthe study. T h e authors conclude that being identified as a false-positive participant is clearly not benign (Gotlib et al., 1995). Two studies including diagnostic interviews with parents and parent reports on the CBCL have found considerable correspondence between the two approaches. Edelbrock and Costello (1988a), in a study of clinically referred children and adolescents (ages 6 through 1 6 years), found that scores o n the Depressed Withdrawal scale ofthe CBCL (derived from the initial version of the CBCL profile) for girls aged 12-16 correlated significantly with diagnoses of M D D and DD, and that scores on the Uncommunicative scale of the

CBCL for boys ages 6-16 correlated with DSMI11 diagnoses of Overanxious Disorder of Childhood, M D D , and DD. T h e broad-band Internalizing scale and Externalizing scale were also correlated with diagnoses of M D D (. 3 1, p < .001, and .14, p < .01, respectively) and D D (.43, p < ,001, and .22, P < .001, respectively). They also found that scores on the Depression scale for children aged 6-1 l were linearly related to the probability of receiving a diagnosis of either M D D or D D . This finding suggests that there was not a specific threshold score of symptoms on this scale above which children received a depressive diagnosis and below which they did not. It will be important to determine whether a similar pattern is found for adolescents. In a study of 667 clinically referred Australian adolescents, Rey and Morris-Yates (1991) generated DSM-I11 diagnoses on the basis of clinical interviews and calculated scores on a Depression scale ofthe CBCL (identified by Nurcombe et al., 1989). T o examine the correspondence between diagnoses and CBCL scale scores, these authors used receiver operating characteristic (ROC) techniques developed in signal detection theory as an overall index of diagnostic accuracy (in terms of sensitivity and specificity) in analyses of continuous scores. Individuals with a diagnosis of M D D (n = 23) were compared in separate analyses with individuals with all other diagnoses (n = 634), individuals with a diagnosis of DD (n = 62), and individuals with a diagnosis of Separation Anxiety Disorder (n = 57). Those with a diagnosis of M D D scored significantly higher on the CBCL Depression scale than those in any of the other diagnostic groups. Furthermore, ROC analyses indicated that the CBCL scale functioned at a level better than chance in discriminating the M D D group from each of the other three diagnostic groups. Rates ofsensitivity (83%) and specificity (55%) were better in distinguishing M D D from all other diagnoses than in distinguishing M D D from D D or Separation Anxiety Disorder. Additional analyses with this sample compared the sensitivity and specificity of six depression scales extracted from the CBCL and the YSR (Rey & Morris-Yates, 1992). All of the scales performed at a level better than chance in R O C analyses discriminating adolescents with M D D diagnoses from those without. Using adolescents7self-reports on the YSR and the child version of the DISC, Weinstein et al. (1990) also found evidence for a significant positive association between DSM-111 diagnoses of affective disorders and scores on the YSR De-

216

P A R T 111. E M O T I O N A L A N D S O C I A L D I S O R D E R S

pressed scale (derived from the initial version of the YSRprofile). However, adolescents who were diagnosed with M D D andlor D D scored higher than those without the diagnosis of an affective disorder on all narrow-band scales on the YSR (Depressed, Unpopular, Somatic Complaints, Aggressive, Delinquent, Thought Disorder) and on the broader scales of Internalizing and Externalizing problems. Finally, Gerhardt et al. (1996) examined the correspondence among depressed mood, the Anxious/Depressed syndrome, and an analogue measure of M D D in parent and adolescent reports on the CBCL and YSR. Adolescents' self-reports on theYSR indicated that of those who met the DSM analogue criteria for MDD, 59% were also in the clinical range on the AnxiousDepressed syndrome; of those who were in the clinical range on the Anxious/Depressed syndrome, 17% met the criteria for the analogue of NIDD. The pattern was similar for parents' reports on the CBCL: 71% of adolescents who met the DSM analogue criteria also were in the clinical range on the Anxious1 Depressed syndrome; 12% of those who were in the clinical range on the syndrome also met the criteria for the analogue of MDD. Relative-risk odds ratios indicated that youths who met criteria for the Anxious/Depressed syndrome at one assessment were 9.3 times more likely based on parent reports and 15.5 times more likely based on selfreports to meet criteria on the analogue of M D D 3 years later (Gerhardt et al., 1996). Thus, the Anxious/Depressed syndrome identified a larger group of children and youths, a subset of whom also met criteria associated with clinical depression. Asubstantial portion of those who met criteria for MDD, however, were not in the clinical range on the AnxiousDepressed syndrome. Those children and adolescents who exceeded criteria on both indices were significantly more impaired on a number of measures than those who only exceeded either the syndrome cutoff or the M D D cutoff (Gerhardt et al., 1996). These studies indicate that diagnoses of M D D and D D derived from clinical interviews are related, albeit imperfectly, both to scores on selfreport inventories of depressive symptoms (including depressed mood) and to depressive syndrome scores from multivariate checklists. We (Compas et al., 1993) have proposed a hierarchical and sequential model to describe the association among depressed mood (as measured by selfreport questionnaires), depressive syndromes (as measured by behavior checklists), and depressive

disorders (as measured by diagnostic interviews). These three levels of depressive phenomena are hypothesized to reflect progressively severe manifestations of depressive problems, with depressed mood functioning as a risk for the development of the syndrome, and the syndrome functioning as a risk factor for the disorder. The data described above are generally consistent with this model, although the overlap of the three levels of depressive problems is imperfect. Moreover, it is important to be aware that children/adolescents who obtain high depressive symptom scores on selfreport questionnaires, who score in the clinical range on the AnxiousDepressed syndrome, or who meet DSM-IV criteria for M D D or D D all represent significant clinical problems. That is, youths in all of these groups are experiencing significant levels of emotional distress and substantial levels of impairment in social functioning (Gerhardt et al., 1996; Gotlib et al., 1995).

ASSESSMENT OF CONSTRUCTS RELATED T O DEPRESSION

The biopsychosocial perspective on depression in young people (Petersen et al., 1993) emphasizes the need to consider a range of important correlates of depression. Assessment of only the core symptoms of a depressive syndrome or disorder will offer an inadequate understanding of a child's or adolescent's functioning, the potential causes of the problem, co-occurring or comorbid problems, and associated factors that could serve as the target for interventions. The number of associated factors is vast, however, and a thorough review of the assessment ofthese factors is beyond the scope of this chapter. An overview of these factors and their measurement, although limited, is instructive to guide a more comprehensive approach to the assessment of depression.

Biological Factors
Current evidence ~ o i n tto the central role of neus roendocrine functioning in depression (BrooksGunn, Petersen, & Compas, 1995; Shelton, Hollon, Purdon, & Loosen, 1991). The vegetative symptoms of depression (e.g., sleep and appetite disturbances), as well as mood disturbances, are believed to be related to dysregulation of the limbic system. T h e limbic system, involved in the regulation of drive. instinct. and emotions. has bgen studied through two' key pathways: the

Chapter 4. Depression in Children and Adokscents

21 7

hypothalamic-pituitary-adrenal (HPA) axis and the hypothalamic-pituitary-thyroid ( H I T ) axis. In spite of significant advances in research on the biological correlates of depression in children and adolescents, there are no standardized biological measures that have proven useful in identifying depression in ~ o u n people. Much of the adult g and childladolescent research on biological correlates of depression has focused on the study of the HPA axis. The functioning of the HPA axis is monitored by measurement of the level of cortisol released in response to a known suppressor, dexamethasone. Cortisol, a substance released by the adrenal glands in response to hormones released by the hypothalamus and pituitary, is described as preparing the organs for physical action. Through the dexamethasone suppression test (DST), depressed adults have been identified as having abnormally high levels of cortisol in the presence of dexamethasone (for a review, see Arana, Baldessarini, & Ornstein, 1985).The DST has been found to have an overall sensitivity of 45% and specificity of96% with depressed adults. However, these rates are lower in children and adolescents (Kutcher et al., 1991; Hughes &Preskorn, 1989). For example, in one study of adolescents, only 40% of the depressed subjects and 69% of the normal subjects were identified correctly with the DST (Kutcher et al., 1991). Researchers of child/adolescent depression have concluded that the DST has not yet been shown to be a reliable or valid diagnostic tool for children or adolescents. The difference between childladolescent depressives' and adult depressives' cortisol responses on the DST is thought to be a result of possible maturational differences of the neuroendocrine systems (Dahl et al., 1989, as cited in Kutcher et al., 1991).Additional explanations of the age difference in DST results offered by Dahl and associates (1989, as cited in Kutcher et al., 1991) include the following: (1)There may be age differences in the effects of chronic stress responses on cortisol patterns in children and adolescents; (2) adult depression may have had longer to affect the central nervous system; and (3) a history of adult antidepressive medication may affect adult neuroendocrine responsivity. The other neuroendocrine pathway studied in relation to depression, the HPT axis, involves measurement of the level of thyrotropin-releasing hormone (TRH). TRH, believed to be associated with an increased sense of relaxation, activity, and positive mood (Garcia et al., 1991), stimulates the production of thyroid-stimulatinghormone (TSH)

in the pituitary gland. In depressed adults, administration of TRH causes "blunted or decreased" production and release of TSH (Garcia et al., 1991). Studies with de~ressed childrenladolescents'and controls matAed for age and sex have failed to find consistent significant differences in the TSH levels in response to TRH (Garcia et al., 1991: Kutcher et al.. 1991). The neuroendocrine system's functioning in depressed children and adolescents has been further examined through levels of growth hormone. Whereas the evidence for hypersecretion of growth hormone in depressed adults remains conflicting, ". nocturnal secretions of growth hormone were found to be significantly higher in depressed adolescents (Kutcher et al.. 1991).The measurement of nocturnal growth hdrmonk showed 99% sensitivity and 93% specificity with respect to the identification of depressed adolescents. Nonetheless, the researchers suggested that more study of 24-hour differences in growth hormone levels in depressed adolescents required, because of the complexity of growth hormone secretion. It is, however. the onlv neuroendocrine marker that reliably distinguished depressed adolescents from their normal peers. Finally, disruptions of biological rhythms have been studied through the monitoring of sleep patterns. Research on adult devressives indicated that there are sleep continuity histurbances, reduced delta (slow-wave)sleep, high rapid-eye-movement (REM) density, and shortened REM latency (onset of REM after sleep begins) (see Emslie, Rush, Weinberg, Rintelmann, & Roffwarg, 1990). Although Emslie et al. (1990) cited evidence of some sleep disruptions among depressed adolescents (e.g., shortened REM latency, REM density and sleep continuity problems), they concluded that the findings are still too conflicting to use polysomnographic measures as a diagnostic tool of childladolescent de~ression. In sum, neuroendocrine dvsfunction and biological rhythm disruptions ap;>earto be related to childladolesce~lt depression, but at this time they cannot be reliably used as diagnostic measures of the presence, severity, and type of depressive symp toms. Differences in growth hormone levels and sleep patterns show some promise in terms of being able to distinguish depressed children or adolescents from their peers. It is unclear whether these biological also distinguish between depressed childrenladolescents and other ~ s ~ c h i atric groups (e.g., those with Conduct Disorder vs. those with MDD). Furthermore, all of the neu-

218

P A R T 111. E M O T I O N A L A N D S O C I A L D I S O R D E R S

roendocrine assessment prbcedures require insertion of a catheter or needle to make frequent blood withdrawals over a 24-hour period. T h e invasive nature of these procedures raises serious questions about their use in outpatient settings.

Social-Cognitive Factors
Extensive empirical evidence now links a number of social-cognitive factors to levels of depressive symptoms in children and adolescents. These factors include self-schemas, cognitive errors or distortions, views ofthe self, hopelessness, perceived control, and attributional or explanatory style (see Kaslow, Brown, & Mee, 1994, for a review). With regard to each of these constructs, evidence suggests that depressive symptoms in children and adolescents are associated with more negative views of the self, the social world, and the future. T h e quality and level of standardization of measures of each of these domains vary considerably, however. As a result, some of these aspects of social-cognitive functioningare better suited than others for assessment in clinical contexts. Several studies have indicated that higher levels of depressive symptoms are associated with more negative self-schemas. In both incidental recall ofself-descriptive traits (Hammen & Zupan, 1984; Prieto, Cole, & Tageson, 1992) and recall of words from a word association task (Prieto et al., 1992; Whitman & Leitenberg, 1990), children with higher levels of depressive symptoms recalled more negative words. This suggests that depressive symptoms are associated with more negative cog-, nitive structures, especially those that reflect representations ofthe self. Standardized measures of cognitive schemas that can be used in clinical contexts are not available, however. Drawing on Beck's cognitive model of depression in adults, several investigators have examined cognitive errors and distortions that may be associated with depressive symptoms in children and adolescents. Measures of children's cognitive errors include the Cognitive Bias Questionnaire for Children (Haley, Fine, Marriage, Moretti, & Freeman, 1985), the Children's Negative Cognitive Error Questionnaire (Leitenberg et al., 1986), and the Automatic Thoughts Questionnaire for children (Kazdin, 1990).These measures are all selfreport questionnaires that present respondents with either hypothetical situations (Leitenberg et al., 1986) or a list of statements (Kazdin, 1990) reflecting negative cognitive distortions about the self, the world, or the future. T h e reliability and validity of these scales have been established, and

they have been constructed specifically for use with children and adolescents (i.e., they are not simply downward extensions of adult measures). Negative views of the self are central to cognitive models of depression in young people, and several measures are available to assess various aspects of children's self-concept, self-esteem, or perceptions of personal competence. T h e most comprehensive and widely used of these scales is the Self-Perception Profile for Children (SPPC; Harter, 1985) and the parallel form for adolescents (Harter, 1988). The SPPC is noteworthy in several regards. First, it was designed to assess multiple dimensions ofthe self, including perceptions of academic competence, social relations, physical appearance, athletic competence, behavioral conduct, and global self-worth. This yields a more detailed and differentiated profile of children's self-perceptions than can be obtained from other measures of self-esteem or self-concept. Second, the format of the scale is designed to reduce social desirability in responding. Third, the psychometric properties of the scale have been well established. And fourth, the SPPC has been used in a number of studies and has been shown to be related to depressive symptoms in children and adolescents (e.g., Harter & Marold, 1992; Robinson, Garber, & Hilsman, 1995). Perceptions of personal control play an important role in several models of depression, and have been examined in studies of depressive symptoms in children. This research suggests that simple, unidimensional conceptualizations of control along an internal-external dimension are inadequate; more complete models indicate that control beliefs are the result of judgments of contingencies and personal competence (Skinner, 1995; Weisz, 1986). For example, Weisz and colleagues have shown that depressive symptoms are predicted more strongly by beliefs about competence and control than by beliefs about contingencies (Weisz, Weiss, Wasserman, & Rintoul, 1987), although a recent study found associations of depressive symptoms with both competence and contingency beliefs (Weisz et al., 1993). Contingency and competence beliefs were measured most successfully by a set of simple probes to assess each of these constructs-probes that can be used easily in both clinical and research contexts (Weisz et al., 1987). Finally, the most extensive research on socialcognitive processes in child and adolescent depression has been based on the model oflearned helplessness. This has involved measurement of what has been variously labeled "attributional style,"

Chapter 4 Depress~onin Chrldren and Adolescents

2 19

"explanatory style," or "pessimism." Regardless of which label is applied, measures have focused on the ways in which children and adolescents perceive the causes of success and failure in their lives. The most widely used measure of attributional style is the ChiIdren's Attributional Style Questionnaire (CASQ; Seligman et al., 1984). The CASQ consists of a series of hypothetical success and failure situations; the respondent is asked to choose between two attributions for the cause of each event. The responses are forced choices that vary the dimensions of internal-external, stable-unstable, and global-specific causes. The psychometric properties of the CASQ are estabIished. A number of cross-sectional and prospective studies have used the CASQ and found a relationship between a maladaptive attributional styIe and depressive symptoms. Moreover, attributions interact with other factors, including stressful life events and self-perceptions of competence, in predicting depressive symptoms (e.g., Hilsman & Garber, 1995; Robinson et al., 1995; Turner & Cole, 1994). Furthermore, the findings ofNolenHoeksema, Girgus, and Seligman (1992) indicate that the association between attributions and depressive symptoms may change with development. Behavioral and Interpersonal Factors Deficits in interpersonal skills, problem-solving skills, and coping strategies are all implicated in depression in children and adolescents. Higher depressive symptoms are associated with poorer interpersonal competence (Cole, 1990,1991; Cole &White, 1993), inadequate problem-solving skills (Rudolph, Hammen, & Burge, 1994), and a coping style associated with poor regulation or management of emotions (Carber, Braafladt, & Weiss, 1995). Unfortunately, many of the measures of these constructs are not easily adapted for use in current clinicaI practice, as they either involve peer nominations (which are difficult to obtain in clinical settings) or are measures with unknown or inadequate psychometric qualities. Family Factors The primary family characteristics that have been implicated in depression in young people are the presence of depressive symptoms or a depressive disorder in a mother or father (Hammen, 1991; Phares & Compas, 1992). Specifically, a history of depressive disorder in parents is a well-established risk factor for emotional and behavioral problems in children and adolescents, including

(but not limited to) increased risk for depressive symptoms and disorders (Hammen, 1991). In addition to a prior history of mood disorder, current levels of depressive symptoms in mothers have been shown to constitute a powerful predictor of depressive symptoms in children (e.g., Hammen et al., 1987; Hammen, Burge, & Adrian, 1991). These findings indicate that comprehensive assessment of depression in children and adolescents will be informed by the assessment of current depressive symptoms and prior history of depressive disorder in parents. This can be accomplished through the use of symptom inventories such as the BDI or the CES-D, and the use of structured diagnostic inventories designed for adults. Other aspects of family functioning have been implicated in childhood depression, but the evidence and measures are not sufficiently well established to warrant use in clinical practice at this time. For example, childhood depression is associated with negative parent-child interactions; family environments that are perceived as less cohesive, less open to emotional expressiveness, and more hostile and rejecting; and more stressful family environments (see Hammen & Rudolph, 1996). Although these factors remain high priorities for future research, their application in clinical contexts appears to be premature at this time. Environmental Factors In addition to the family, other aspects of the social environment have been implicated as correlates of depressive symptoms in children and adolescents. Foremost among these is the occurrence of stressful major life events, as well as exposure to ongoing stresses and strains (sometimes referred to as "minor events," "daily hassles," or "recurrent strains"). Several prospective longitudinal studies have shown that stressors, particularly minor events or hassles, are predictive of subsequent depressive symptoms even after initial depressive symptoms are controlled for (e.g., Compas, Howell, Phares, Williams, & Ciunta, 1989; DuBois, Felner, Brand, Adan, & Evans, 1992; Nolen-Hoeksema et al., 1992). That is, stress predicts increased depressive symptoms over time, but depressive symptoms also predict increased minor or chronic stress (Adrian & Hammen, 1993; Compas et al., 1989). The measurement of major and minor stressful events in the lives of children and adolescents has been accomplished through questionnaires completed by parents (e.g., Coddington, 1972), questionnaires completed by children or adoles-

220

P A R T 111. E M O T I O N A L A N D S O C I A L D I S O R D E R S

cents (e.g., Compas, Davis, Forsythe, &Wagner, 1987), and structured interviews with parents and/ or children and adolescents (e.g, Hammen, 1988). Although these procedures are useful in generating detailed information about current sources of stress in the lives of children and adolescents, most are quite long and thus too cumbersome to be used in clinical practice.

Cornorbid Disorders and Co-Occurring Problems

Symptoms, syndromes, or disorders rarely occur alone during childhood and adolescence. Nowhere is this phenomenon, referred to as the "covariation" or "co-occurrence" of symptoms and the "comorbidity" of disorders, more evident than for depressive problems in children and youths (Angold & Costello, 1993; Compas & Hammen, Several studies have shown that during childhood and adolescence, as in adulthood, depressed mood is closely related to other negative emotions. First, monomethod studies (i.e., studies relying on a single method, such as child/adolescent selfreports) have failed to distinguish depressed mood (and other symptoms of depression) from other negative emotions, including anxiety, anger, and hostility. For example, Saylor et al. (1984) found that children and adolescents classified as high and low in depressive symptoms on the CDI also differed significantly on self-reported anxiety. Furthermore, multitrait-multimethod validity studies examining reports from different informants (e.g., children/adolescents, teachers, parents) of various negative emotions (depression, anxiety, anger) have found that strong associations between youths' depressed mood and other negative emotions, especially anxiety, are not limited to youths' self-reports of emotions. That is, reports ofyouths' depressed mood by one group of informants are correlated more highly with reports of other negative emotions by those same informants than they are correlated with reports of depressed mood obtained by other groups of informants (e.g., Wolfe et al., 1987). These findings are consistent with the results ofprincipal-component analyses of the CBCL, YSR, and TRF, which revealed the syndrome of mixed anxiety and depression symptoms discussed earlier. It is noteworthy that parent and teacher reports of youths' emotions and behaviors also show considerable covariance in levels of depressed and anxious mood (Finch et a]., 1989). Thus, the association ofdepressed mood with other elements

of the broader construct of negative affect (see below) is not the result of a simple bias in young people's reports about their internal emotional states. Finch et al. (1989) suggest that anxiety and depression are not separable in children and adolescents, and that the distinction between these two forms ofnegative affect should be put to rest. These findings must be interpreted with a level of caution, however, as there is some degree of item contamination between measures of depressed and anxious mood. For example, Brady and Kendall (1992) have identified several items that are included on both the CDI and standard self-report scales of anxiety. The extent to which item similarity on these measures accounts for the degree of association between the scales has not been clarified. Concerns about confounding of measures notwithstanding, these findings can be understood by considering them within the broader framework of theories of emotion (e.g., Watson & Tellegen, 1985). Extensive evidence from studies of the structure of emotions in children, adolescents, college students, and adults indicates that self-rated mood is dominated by two broad factors: "negative affect," which is composed of negative emotions and distress, and "positive affect," which is made up of positive emotions (King et a]., 1991; Watson, 1988; Watson & Tellegeri, 1985). Depressed mood is one component of the broader construct of negative affect, whereas positive emotions are important in distinguishing among subtypes of negative emotion (Watson & Clark, 1984). Research with adults indicates that although depressed mood is strongly intercorrelated with other negative emotions, it appears to be distinguishable from anxiety, if not other forms of negative affect, on the basis of its association with positive emotions (e.g., happiness, excitement, pride, contentment). Specifically, whereas anxiety is uncorrelated with positive affect, depressed mood shows a consistent inverse relationship with positive affect (Watson & Kendall, 1989).Thus, highly anxious individuals may be low, moderate, or high in positive affect, as anxiety and positive emotions can co-occur. In contrast, highly depressed individuals are likely to experience low levels of positive enlotions (i.e., anhedonia). The relation between positive affect and depressed mood during adolescence warrants further research. In general, research suggests that sad or depressed mood is a phenomenologically distinct emotional state that, although closely related to the experience of other forms of negative affect, is distinguished by its relation to positive affect (Watson & Clark, 1992).

Chapter 4 Depression In Ch~ldren Adolescents and

22 1

The intercorrelabons of the Anx~ous/De~ressed core syndrome with the other core syndromes indicate substantial covanation. These correlations have been reported separately for the CBCL,TRF, and YSR for clin~cally referred and nonreferred adolescent boys and girls (Achenbach, 1991a). Although these correlations vary substantially, ranging from .27 (with Delinquent for referred boys on the YSR) to .80 (with Self-Destructive for referred boys on theYSR), the overall mean correlation of the Anxious/Depressed syndrome with the other core syndromes is .5 1, indicating substantial covariance. Furthermore, the Anxious/ Depressed syndrome correlated highly with both internalizing syndromes (Withdrawn, Somatic Complaints) and externalizing syndromes (Aggressive Behavior, Attention Problems). Thus, the degree of covariation of the Anxious/Depressed syndrome with other syndromes is substantial, The covariation of the Anxious/Depressed syndrome with other syndromes remains high after measurement factors are controlled for (Hinden, Compas, Achenbach, & Howell, in press). Further research is needed to understand the implications of this high degree of covariation for the etiology, course, treatment, and prevention of depressive syndromes (Compas & Hammen, 1994). The comorbidity of MDD with other psychiatric diagnoses has been examined in epidemiological studies of nonreferred samples of children/ adolescents in the community. Community samples rather than clinical samples are necessary to determine true rates of comorbidity, as rates of comorbidity in clinical samples will be disproportionately high because of a number of factors (Caron & Rutter, 1991). Comorbidity appears to be the rule for child/adolescent depression (for reviews, see Brady & Kendall, 1992; Compas & Hammen, 1994; Angold & Costello, 1993). The recent community studies by Lewinsohn and colleagues described above are illustrative in this regard (Lewinsohn et al., 1991; Rohde et al., 199 1). The current comorbidity rate for MDD and D D was 20 times greater than that expected by chance, and the lifetime comorbidity rate was three times greater than chance (Lewinsohn et al., 1991). Levels of comorbidity with disorders other than mood disorders were also high, as 43% of adolescents with a depressive disorder received at least one additional diagnosis-a rate that was 9.5 times greater than chance (Rohde et al., 1991). Current comorbidity was highest for anxiety disorders (18%),substance use disorders (l4%),and disruptive behavior disorders (8%).

IMPLICATIONS FOR ASSESSMENT AND INTERVENTION

Given the possible relations, as well as differences, among levels of depressive phenomena, how can researchers and clinicians select from among the myriad of methods for assessing child and adolescent depression? More specifically, how is one to decide among self-report questionnaires, behavior checklists, and diagnostic interviews? The following criteria are offered to guide the choice of measures of depression in children and adolescents. 1. The measure must reflect a specific diagnostic or classification scheme. That is, the researcher or practitioner must readily acknowledge the taxonomic paradigm that is guiding his or her work and select a measure commensurate with that paradigm. For example, it is widely recognized that self-report measures such as the CDI are adequate measures of depressive symptoms, but do not provide adequate information on the severity and duration of the full complement of depressivesymptoms to enable the clinician to make a diagnostic judgment. As indicated above, the various assessment methods are linked to specific taxonomic systems. 2. The measure must be psychometrically sound. Most of the measures that have been reviewed meet the minimal criteria with regard to internalconsistency reliability and test-retest reliability. However, test-retest reliability of measures of depressed affect may differ substantially for clinically referred and nonreferred populations (e.g., Saylor et al., 1984). As indicated above, validity has proven a more difficult issue to address. Some measures have also established construct validity through factor analysis or principal-component analysis of their underlying structure. Establishing criterion validity has proven much more difficult for each of the types of measures that we have reviewed, in part because ofthe low rates of agreement across different informants. In spite ofthese limitations, the majority of the instruments described in this chapter meet the minimum criteria for use in research and practice. 3. The measure must be developmentally appropriate for use with children and adolescents. The content and wording ofthe items must have been generated for use with this age group. If this is not the case, the measure may reflect the error of making either a downward extension of an adult measure or an upward extension of a child measure. UnfortunateIy, most of the assessment

222

P A R T 111. E M O T I O N A L A N D S O C I A L D I S O R D E R S

procedures for childladolescent depression fail to meet this criterion. For example, the BDI has been extended to research with adolescents without any changes in the content or wording of the items to make them suitable for this age group. Further refinement of some measures may be needed to insure that they are developmentally appropriate for adolescents. 4. The measure must provide sufficient epidemiological data to developmental and normative data comparisons. This is especially important if a cutoff or criterion score is to be used to distinguish between clinical and nonclinical groups. Behavior checklists have been used to generate extenswe, nationally representative normatwe data bases. As one might expect, norms for scales of depressed mood are much more limited, since these measures were not designed for use in making categorical decisions. Diagnostic judgments usingclinical interviews are made independently of normative data on childrenladolescents and rely instead on clinical criteria. However, interpretation ofthese criteria will be strengthened as larger normative data sets are generated through clinical interviews with community samples. 5. If one adheres to a categorical diagnostic model, the measure must provide criteria for making distinctions between clinical disorders a n d subclinical phenomena. Furthermore, it is important that these criteria not be solely a function of the judgment ofthe individual clinician. DSM-N criteria and the RDC have been used as the basis for most diagnostic decisions with the K-SADS, CAS, DISC, and other structured interviews. These criteria appear to be adequate, as acceptable levels of interrater agreement have been generated for these interviews. 6. The measure must be sufficiently sensitive to change for use in the assess ofprevention and treatment effects. The first step in determining sensitivity to change is establishing relatively high testretest reliability over very short periods oftime, to rule out effects of random error in responses or large fluctuations in scores as a result of highly state-dependent factors. In addition, data must be provided to indicate that scores tend to be highly stable over longer periods of time in the absence of any intervention or other major perturbation. The field could benefit from more longitudinal data obtained on all of the measures described

designed to assess only specified symptoms of depression. Although these measures may be sens~tive depressive symptoms, they may lead the to researcher or clinician to the mistaken conclusion that symptomatic elevations are limited to a particular depressive syndrome or disorder. Possible covariation with other symptoms or comorbidity with other disorders will be overlooked. 8. Measures of depressive symptoms or disorders need to be complemented by measures of related domains ofbnctioning, including social-cognitive processes, parental depressive symptoms, and interpersonal competence. Data obtained from these measures will play an important role in the planning, implementation, and evaluation of interventions to address depressive problems in children and adolescents. The selection of measures of other constructs should be guided by one's conceptual framework and by data that support the utility of these constructs in understanding the nature and course of depressive problems in young people. As noted above, measures of a number of these constructs (e.g., stressful life events) are not currently well suited for use in clinical practice, to but are likely to cont~nue play an important role in research. With these criteria in mind, it is useful to consider a "multiple-gating" procedure in clinical practice and research, as suggested by several investigators (e.g., Roberts et al., 1991). However, because of the need to attend to the high rates of covariance of depressive symptoms with other symptoms and comorbidity of depressive syndromes and disorders with other disorders, the procedures recommended here are slightly different from those outlined by others. As a first step, broad-band checklists should be used to assess a wide range of symptoms and problems, including (but not limited to) depressive symptoms. These checklists should include the perspectives of childrenladolescents, parents, and teachers as the primary interested parties in identifying child1 adolescent depression. T h e integrative package of the YSR, CBCL, and T R F offers the most integrated set of measures to accomplish this goal (Achenbach, 1991a).Along with these broad-band measures, a more focused measure of depressive symptorns should be used to target the emotions associated with depressive disorders. Ofthe many measures that are available, the CDI has shown the greatest strengths for assessing recent depressive symptoms. The CDI should be administered on at least two occasions over a 2-week period to determine the persistent versus transient nature of

7. The measure must be sufficiently broad in focus to allow for determination of co-occurrence ofothersymptoms orcomorbidity ofotherdisorders. This is a limitation of many measures that are

Chapter 4. Depression in Children and Adolescents

223

these symptoms. These four measures (the YSR, CBCL, TRF, and C D I ) can be used to identify individuals who exceed clinical criteria on the Anxious/Depressed syndrome on the multivariate scales and on the total score on the CDI. In addition to examining scores on the AnxiouslDepressed syndrome, the clinician can accomplish an initial screening for the presence of clinical elevations on other problems by examiningscores on the other seven core syndromes on the YSR, CBCL, and TRF. In summary, the CDI, YSR, CBCL, and T R F should be used to screen for clinically significant elevations in negative affect, the core Anxious/Depressed syndrome, and other clinical syndromes that could indicate the presence of other internalizing andlor externalizing problems. Individuals who exceed the clinical cutoff on the CDI or the AnxiousIDepressed syndrome on one or more ofthe checklists should then be designated for further assessment through a diagnostic interview to determine the presence of diagnosable depressive disorders. Through this sequence of assessment procedures, distinct subgroups of individuals could be identified who warrant different intervention procedures.
REFERENCES
Achenbach, T . M. (1985). Assessment and taxonomy of child and adolescent psychopathology. Beverly Hills, CA: Sage. Achenbach, T . M. (I991a). Integrative guide for the 1991 CBCLI4-18, YSR, and TRF Profiles. Burlington: University of Vermont, Department of Psychiatry. Achenbach, T . M. (199 I b). Manual for the Child Behavior Checklist a n d 1991 Profile. Burlington: Universityofvermont, Departmentofpsychiatry. . Achenbach, T. M. ( 1 9 9 1 ~ )Manual for the Teacher's Report Form and 1991 Profile. Burlington: University of Vermont, Department of Psychiatry. Achenbach, T . M. (199ld).Manualfor the Youth SelfReport and 1991 Profile. Burlington: University of Vermont, Department of Psychiatry. Achenbach, T. M. (1993). Empirically based taxonomy. Burlington: University of Vermont, Department of Psychiatry. Achenbach, T. M., Conners, C . K., Quay, H. C., Verhulst, F. C., & Howell, C . T. (1989). Replication of empirically derived syndromes as a basis for a taxonomy of child/adolescent psychopathology. Iournal of Abnormal Child Psychology, 17, 299-323. Achenbach, T. M., & Edelbrock, C. (1981). Behavioral problems and competencies reported by parents

of normal and disturbed chiIdren aged four to sixteen. Monographs of the Society for Research in Child Development, 46(1, Serial No. 188). Achenbach, T. M., C. T., Quay, H. C., & Conners, C . K. (1991). National survey of problems and competencies among four- to sixteenyear-olds: Parents' reports for normative and clinical samples. Monographs of the Society for Research in Child Development, 56(Whole No. 3). Achenbach, T . M., McConaughy, S. H., & Howell, C . T. (1987). Child/adolescent behavioral and emotional problems: Implications of cross-informant correlations for situational specificity. Psychological Bulletin, 101, 2 13-232. Adrian, C., & Hammen, C. (1993). Stress exposure and stress generation in children of depressed mothers. Ioumal of Consulting and Clinical Psychology, 61, 354-359. ~ , psychiatric ~~ ~ i ~ ~ ~ ~ ~ ~ (1994). ~ ii ~ ~ and statistical manual of mental disorders (4th ed.). Washington, DC: Author. Angold, A. (1988). Childhood and adolescent depression: 1. Epidemiological and aetiological aspects. British \oumal of Psychiatry, 152, 601-617. A"g"'d> A.7 Costello, E. J. (1993). Depressive comorbidity in children and adolescents: Empirical, theoretical, and methodological issues. American Ioumal of Psychiatry, 1SO, 1779-1791. Angold, A., & Rutter, M. (1992). Effects of age and pubertal status on depression in a large clinical sample. Development and Psychopathology, 4, 5-28. Arana, C . W., Baldessarini, R. J., & Omstein, M. (1985). The dexamethasone suppression test for diagnosis and prognosis in psychiatry. Archives of General Psychiatry, 42, 1193-1204. Archer, R. P., Pancoast, D. L., & Klinefelter, D. (1989). A comparison of MMPl code types produced by traditional and recent adolescent norms. Psychological Assessment: A loumal of Consulting and Clinical Psychology, 1, 23-29. Beck, A. T., Ward, C . H., Mendelson, M., Mock, J. E., & Erbaugh, J. K. (1961). An inventory for measuring depression. Archives of General Psychiatry, 4, 561-571. Bidaut-Russell, M., Reich, W., Cottler, L. B., Robins, L. N.,Compton, W. M., & Mattison, R.E. (1995). The Diagnostic Interview Schedule for Children (PC-DISC v. 3.0): Parents and adolescents suggest reasons for expecting discrepant results. \oumal of Abnormal Child Psychology, 23,641-659. Brady, E. U., &Kendall, P. C. (1992). Comorbidityof anxiety and depression in children and adolescents. Psychological Bulletin, 111, 244-255. Brooks-Cunn, J., Petersen, A. C., & Compas, B. E. (1995). Physiological processes and the development of childhood and adolescent depression. In I. M. Coodyer (Ed.),The depressed child and adolescent: Developmental and clinical perspecfives

~ t

224

P A R T 111. E M O T I O N A L A N D S O C I A I - D I S O R D E R S

(pp. 81-109). New York: Cambridge University Press. Cantwell, D P., &Baker, L. (1991). Manifestations of depress~ve affect in adolescence. Journal ofyouth and Adolescence, 20, 12 1-1 33. Carlson, G A. (1994) Adolescent bipolar d~sorder: Phenomenology and treatment impl~cat~ons. ln W. M. Reynolds & H. F. Johnston (Eds.), Handbook of depression in chtldren and adolescents (pp. 41-60). New York: Plenum Press. Carlson, G. A,, & Cantwell, D. P (1980). Unmasking masked depression in children and adolescents. American Journal of Psychiatry, 137,445-449. Caron, C , & Rutter, M. (1991). Comorbidity in child psychopathology Concepts, issues and research strategies. Journal of Child Psychology and Psychiatry, 32, 1063-1080. Chambers, W. J., Puig-Antich, J., Hirsch, M., Paez, P., Ambrosini, P. J., Tabriz~,M., & Davies, M. (1985). The assessment of affective disorders In chlldren and adolescents by semistructured interview. Archives ofGenera1Psychiatry, 42,696-702. Ciccheth, D ,Rogosch, F A., & Toth, S. L. (1994). A developmental psychopathology perspective on depression ~n children. In W. M. Reynolds & H. F. Johnston (Eds.), Handbook of depression in children and adolescents (pp. 97-122). NewYork. Plenum Press. Clarke, G. N., Lewinsohn, P. M., Hops, H., & Seeley, J. R. (1992). A self- and parent-report measure of adolescent depression: T h e Child Behavlor Checkl~st Depression scale. Behavioral Assessment, 14, 443-463. Coddington, R. D. (1972) The significance of life events as etiologic factors in the diseases of children-1. Journal of Psychosomatic Research, 16, 205-213 Cole, D. A. (1990). Relation of social and academic competence to depressive symptoms in childhood. Journal of Abnormal Psychology, 99,422429. Cole, D. A (1991). Preliminary support for a competency-based model of depression in children Journal ofAbnormal Psychology, 100, 181-190 Cole, D. A,, & Turner, J. E. (1993). Models of cognitive mediation and moderation in child depression. Ioumal of Abnormal Psychology, 102, 271281. Cole, D. A., & White, K. (1993). Structure of peer ~mpressions children's competence: Validation of of the peer nom~nation multiple competencies. of Psychological Assessment, 5,449-456. Compas, B. E , Davis, G E., Forsythe, C J., & Wagner, B. M. (1987). Assessment of major and daily stressful events during adolescence: T h e Adolescent Perce~vedEvents Scale. Journal of Consulting and ClinicalPsychology, 55,534-541. Compas, B. E., Ey, S., & Grant, K E. (1993). Taxonomy, assessment, and diagnosis of depression dur~ng adolescence. Psychological Bulletin, 114, 323-344.

Compas, B. E., Grant, K., & Ey, S. (1994) Psychosoclal stress and child/adolescent depression Can we be more specific? In W. M Reynolds & H F. Johnston (Eds ), Handbook of depress~onIn children and adolescents (pp 509-523) New York, Plenum Press Compas, B. E., & Hammen, C . L (1994). Depression in ch~ldhoodand adolescence: Covarlation and comorbldlty In development. ln R. J. Haggerty, N. Garmezy, M. Rutter, & L. Sherrod (Eds.), Risk and resilience In children: Developmental approaches (pp. 223-267). New York: Cambridge University Press. Compas, B E., Howell, D. C., Phares, V., Williams, R. A,, & Glunta, C . T . (1989). Rlsk factors for emotional-behav~oralproblems in young adolescents: A prospect~ve analysls of adolescent and parent stress. Journal of Consulting and Clinlcal Psychology, 57, 732-740. Compas, B. E., Oppedisano, G., Connor, J. K , Gerhardt, C A,, Hinden, B. R., Achenbach, T. M , & Hammen, C. ( ~ press) Gender differn ences ~n depressive symptoms ~n adolescence: Comparison of national samples of clinicallyreferred and non-referred youth. Journal of Consulting and Clinical Psychology Connor, J. K., Compas, B. E., Gerhardt, C A,, Oppedisano, G., Hinden, B R , & Achenbach, T . M. (1996). Analogue measures of DSM mood and anxiety disorders. Manuscript subm~tted for publication Conners, C. K. (1973). Rating scales for use in drug stud~es with ch~ldren.Psychopharmacology BulIetin: Special Issue, Pharmacotherapy with Children, 24-84. Costello, E. J., Edelbrock, C . S., & Costello, A. J. (1985). Validlty of the NIMH Diagnostic Interview Schedule for Ch~ldren: comparison beA tween psychiatric & pediatric referrals. Journal of Abnormal and C h ~ l d Psychology, 13(4), 579-595. Costello, A. J., Edelbrock, C , Dulcan, M . K., Kalas, R., & Klaric, S. H. (1984). Development and testing of the NIMH Diagnostic Interview Schedule for Children in a clinic population. (Final report, Contract No. RFP-DB-81-0027). Rockville, MD: Center for Ep~demiologlc Studies, National Instltute of Mental Health. Coyne, J. C., & Downey, G. (1991) Social factors and support, and coppsychopathology: Stress, soc~al ing processes. Annual Revlew of Psychology, 42, 401-425. DeCroot, A., Koot, H. M , & Verhulst, F. C. (1994). Cross-cultural generalizabil~t~ the Child Behavof ior Checklist cross-informant syndromes. Psychological Assessment, 7, 299-323 DuBois, D L., Felner, R. D., Brand, S., Adan, A. M., & Evans, E. G. (1992). A prospective study of life stress, social support, and adaptation ~nearly adolescence. Child Development, 63, 542-557. Edelbrock, C ,& Costello, A. J. (1988a). Convergence between statistically der~vedbehavior problem

Chapter 4. Depression in Children and Adohscents syndromes and child psychiatric diagnoses. Journal of Abnormal Child Psychology, 16, 2 19-23 1. Edelbrock, C., & Costello, A. J. (198Sb). Structured psychiatric interviews for children. In M. Rutter, A. H. Tuma, & I. S. Lann (Eds.) Assessment and diagnosis in child psychopathology (pp. 87-1 12). New York: Guilford Press. Edelbrock, C., Costello, A. J., Dulcan, M. K., Kalas, R., & Conover, N. C. (1985). Age differences in the reliabilityofthe psychiatric interview ofthe child. Child Development, 56, 265-275. Emslie, G . J., Rush, J., Weinberg, W. A., Rintelmann, J. W., & Roffwarg, H. P. (1990). Children with major depression show reduced rapid eye movement 1atencies.Archivesof General Psychiatry, 47, 119-124. Fechner-Bates, S., Coyne, J. C., & Schwenk, T . L. (1994). The relationship of self-reported distress to depressive disorders and other psychopathology. Iournal of Consulting and Clinical Psychology, 62, 550-559. Finch, A. J., Lipovsky, J. A., & Casat, C . D. (1989). Anxiety and depression in children &adolescents: Negative affectivity or separate constructs? In P. C . Kendall & D. Watson (Eds.), Anxiety and depression: Distinctive and overlapping features. New York: Academic Press. Fisher, P. W., Shaffer, D., Piacentini, J., Lapkin, J., Kafantaris, V., Leonard, H., & Herzog, D. B. (1993). Sensitivity of the Diagnostic Interview Schedule for Children, 2nd edition (DISC-2.1) for specific diagnoses of children and adolescents. Journal of the American Academy of Child and Adolescent Psychiatry, 32, 666-673. Fleming, J. E., & Offord, D. R. (1990). Epidemiology of childhood depressive disorders: A critical review. Journal of the American Academy of Child and Adolescent Psychiatry, 29, 571-580. Garber, J., Braafladt, N., & Weiss, B. (1995). Affect regulation in children and young adolescents. Development and Psychopathology, 7, 93-1 15. Garcia, M. R., et al. (1991). Thyroid stimulating hormone response to thyrotropin in prepubertal depression. Iournal of the American Academy of Child and Adolescent Psychiatry, 30, 398-406. Garrison, C . Z.,Addy, C . L., Jackson, K. L., McKeown, R. E., & Waller, 1. L. (1991). The CES-D as a screen for depression and other psychiatric disorders in adolescents. Journal of the American Academy of Child and Adolescent Psychiatry, 30,636641. Ge, X., Lorenz, F. O., Conger, R. D., Elder, G. H., & Simons, R. L., (1994). Trajectories of stressful life events and depressive symptonls during adolescence. Developmental Psychology, 30,467483. Gerhardt, C., Compas, B. E., Connor, J. K., Hinden, B. R., & Achenbach, T. M. (1996). Understanding the nature of adolescent depression: Relations among mood, syndrome, and diagnosis. Manuscript submitted for publication.

225

Giaconia, R. M., Reinherz, H. Z., Silverman, A. R., Pakiz, B., Frost,A. K., & Cohen, E. (1993). Ages of onset of psychiatric disorders in a community population of older adolescents. Journal of the American Academy of Child and Adolescent Psychiatry, 33, 706-717. Gotlib, I. H., & Hammen, C. L. (1992). Psychological aspects of depression: Toward a cognitive-interpersonal integration. London: Wiley. Gotlib, I. H., Lewinsohn, P. M., & Seeley, J. R. (1995). Symptoms versus a diagnosis of depression: Differences in psychosocial functioning. Journal of Consulting and Clinical Psychology, 63, 90-100. Haley, G. M. T., Fine, S., Marriage, K., Moretti, M. M., & Freeman, R. J. (1985). Cognitive bias and depression in psychiatrically disturbed children and adolescents. Journal of Consulting and Clinical Psychology, 53, 535-537. Hammen, C. (1 988). Self-cognitions, stressful events, and the prediction of depression in children of depressed mothers. Journal of Abnormal Child Psychology, 16, 347-360. Hammen, C . (1991). Depression runs in families: The social context of risk and resilience in children of depressed mothers. New York: Springer-Verlag. Hammen, C., Adrian, C., Gordon, D., Burge, D., Jaenicke, C., & Hiroto, D. (1987). Children of depressed mothers: Maternal strain and symptoms predictors of dysfunction. Journal of Abnormal Psychology, 96, 190-198. Hammen, C., Burge, D., &Adrian, C . (1991). 'Timing of mother and child depression in a longitudinal study of children at risk. Journal ofConsulting and Clinical Psychology, 59, 341-345. Hammen, C . & Compas, B. E. (1994). Unmasking unmasked depression in children and adolescents: T h e problem of comorbidity. Clinical Psychology Review, 14, 585-603. Hammen, C., & Rudolph, K. D. (1996). Childhood depression. In E. J. Mash & R. A. Barkley (Eds.), Child psychopathology (pp. 153-195). New York: Guilford Press. Hammen, C., & Zupan, B. A. (1984). Self-schemas, depression, and processing of personal information in children. Journal of Experimental Child Psychology, 37, 598-608. Harter, S. (1985). The Self-Perception Profile for Children. Denver, CO: University of Denver, Department of Psychology. Harter, S. (1988). The Self-Perception Profile for Adolescents. Denver, CO: University of Denver, Department of Psychology. Halter, S., & Marold, D. B. (1992).The directionality of the link between selfesteem and affect: Beyond causal modeling. In D. Cicchetti & S. L. Toth (Eds.), Rochester Symposium on Developmental Psychopathology: Vol. 5. The self and its disorders (pp. 333369). Rochester, NY: University of Rochester Press. Herjanic, B., Herjanic, M., Brown, F., & Wheatt, T. (1975). Are children reliable reporters?Journal of Abnormal Child Psychology, 3, 41-48.

226

P A R T 111. E M O T I O N A L A N D S O C I A L D I S O R D E R S

Hilsman, R., & Garber, J. (1995). A test of the cognitive diathesis-stress model of depression in children: Academic stressors, attributional style, perceived competence, and control. lournal of Personality and Social Psychology, 69, 370-380. Hinden, B. R., Compas, B. E., Achenbach, T . M., Hammen, C., Oppedisano, G., Connor, J. K., & Gerhardt, C . A. (1996). Charting the course of depressive symptoms during adolescence: D o we have the right map? Manuscript submitted for publication. Hinden, B. R., Compas, B. E., Achenbach, T. M., & Howell, D. C . (1997). Comorbidity of depression during adolescence: Separating fact from artifact. Journal of Consulting and Clinical Psychology, 65, 6-14. Hodges, K. H. (1990). Depression and anxiety in children: A comparison of self-report questionnaires to clinical interview. Psychological Assessment: A Iournal of Consulting and Clinical Psychology, 2, 376-381. Hodges, K. H., & Cools, J. N. (1990). Structured diagnostic interviews. In A. M. La Greca (Ed.), Through the eyes ofthe child (pp. 109-149). Boston: Allyn & Bacon. Hodges, K. H., Cools, J. N., & McKnew, D. (1989). Test-retest reliability of a clinical research interview for children: The Child Assessment Schedule (CAS). Psychological Assessment: A loumal of Consulting and Clinical Psychology, 1, 317-322. Hodges, K. H., & Craighead, W. E. (1990). Relationship of Children's Depression Inventory factors to diagnosed depression. Psychological Assessment: A loumal of Consulting and Clinical Psychology, 2,489-492. Hodges, K. H., Gordon, Y., & Lennon, M. (1990). Parent-child agreement on symptoms assessed via a clinical research interview for children: T h e Child Assessment Schedule (CAS). Journal of Child Psychology and Psychiatry, 3 1, 427-43 1. Hodges, K. H., Kline, J., Stern, L., Cytryn, L., & McKnew, D. (1982). The developmentofa Child Assessment Schedule for research and clinical use. Journal of Abnormal Child Psychology, 10, 173189. Hodges, K. H., Saunders, W., Kashani, J., Hamlett, K., & Thompson, R. (1990). Internal consistency of DSM-I11 diagnoses using the symptom scales of the Child Assessment Schedule (CAS). Joumal of the American Academy of Child and Adolescent Psychiatry, 29, 635-641. Hughes, C . W., & Preskorn, S. H. (1989). Depressive svndromes in children and adolescents: D i a" n o ~ sis and treatment. Annals of Clinical Psychiatry, 1, 109-1 18. Jensen, P., Roper, M., Fisher, P. W., Piacentini, J., Canino, G., Richters, J., Rubio-Stipec, M., Dulcan, M., Goodman, S., Davies, McRae, D., Shaffer, D., Bird, H., Lahey, B., & Schwab-Stone, M. (1995). Test-retest reliability of the Diagnos- -

tic Interview Schedule for Children (DISC 2.1): Parent, child, and combined algorithms. Archives of General Psychiatry, 52, 61-71. Kandel, D. B., & Davies, M. (1982) Epidelniology of depressive mood in adolescents. Archives of General Psychiatry, 39, 1205-1212. Kashani, J. H., Carlson, G. A., Beck, N. C., Hoeper, E. W., Corcoran, C . M., McAllister, J. A., Fallahi, C., Rosenberg, T. K., & Reid, J. C. (1987). Depression, depressive symptoms, and depressed mood among a community sample of adolescents. American Journal of Psychiatry, 144, 93 1-934. Kashani, J. H., Reid, J. C., & Rosenberg, T. K. (1989). Levels of hopelessness in children and adolescents: A developmental perspective. lournal of Consulting and Clinical Psychology, 57,49&499. Kaslow, N. J., Brown, R. T., & Mee, L. L. (1994). Cognitive and behavioral correlates of childhood depression: A developmental perspective. In W. M. Reynolds & H. F. Johnston (Eds.), Handbook of depression in children and adolescents (pp. 97122). New York: Plenum Press. Kazdin, A. E. (1990). Evaluation of the Automatic Thoughts Questionnaire: Negative cognitive processes and depression among children. Psychological Assessment, 2, 73-79. Kazdin, A. E., Esveldt-Dawson, K., Unis, A. S., & Rancurello, M. D. (1983). Child and parent evaluations of depression and aggression in psychiatric inpatient children. Journal of Abnormal Child Psychology, 11,401-413. Kazdin, A. E., Rogers, A,, & Colbus, D. (1986). T h e Hopelessness Scale for Children: Psychometric characteristics and concurrent validity. Joumal of Consulting a n d Clinical Psychology, 54,241245. Kendall, P. C., Cantwell, D. P., & Kazdin,A. E. (1989). Depression in children and adolescents: Assessment issues and recommendations. Cognitive Therapy and Research, 13, 109-146. King, N. J., Ollendick, T . H., & Gallone, E. (1991). Negative affectivity in children and adolescents: Relations between anxiety and depression. Clinical Psychology Review, 11,441-460. Kovacs, M. (1980). Rating scales to assess depression in school-aged children. Acta Paediatrica, 46, 305-315. Kovacs, M. (1985). The Interview Schedule for Children. Psychopharmacology Bulletin, 21,991-994. Kovacs, M. (1989). Affective disorders in children and adolescents. American Psychologist, 44,209-21 5. Kutcher, S., Malkin, D., Silverberg, J., Marton, P., Williamson, P., Malkin,A., Szalai, J., & Katic, M. (1991). Nocturnal cortisol, thyroid stimulating hormone, and growth hormone secretory profiles in depressed adolescents. Journal of the American Academy of Child and Adolescent Psychiatry, 30(3), 407-414. Leadbeater, B. J., Blatt, S. J., & Quinlan, D. M. (1995). Gender-linked vulnerabilities to depressive symp-

Chapter 4. Depression in Children and Adohscents toms, stress, and ~roblem behaviors in adolescents. Journal of Research on Adolescence, 5, 1-29. Leitenberg, H., Yost, L. W., & Carroll-Wilson, M. (1986). Negative cognitive errors in children: Questionnaire development, normative data, and comparisons between children with and without self-reported symptoms of depression, low self-esteem, and evaluation anxiety. Journal of Consulting and Clinical Psychology, 54, 528-536. Lewinsohn, P. M., Hops, H., Roberts, R. E., Seeley, J. R., & Andrews, J. A. (1993). Adolescent psychopathology: I. Prevalence and incidence of depression and other DSM-111-R disorders in high school students. Journal ofAbnormal Psychology, 102, 133-144. Lewinsohn, P. M., Rohde, P., Seeley, J. R., &Hops, H. (1991). Comorbidity of unipolar depression: I. Major depression with dysthymia. Journal ofAbnormal Psychology, 100, 205-2 13. Lipovsky, J. A,, Finch, A. J., & Belter, R. W. (1989). Assessment of depression in adolescents: Objective and projective measures. Journal ofPersonality Assessment, 53, 449-458. McGee, R., Feehan, M., Williams, S., Partridge, F., Silva, P., &Kelly, J. (1990). DSM-111 disorders in a large sample of adolescents.Journal ofthe American Academy of Child and Adolescent Psychiatry, 29, 61 1-619. Miller, L. C. (1984). Louisville Behavior Checklist manual. Los Angeles: Western Psychological Services. Nolen-Hoeksema, S., & Girgus, J. S. (1994).The emergence of gender differences in depression during adolescence. Psychological Bulletin, 115, 424443. Nolen-Hoeksema, S., Girgus, J. S., & Seligman, M. E. P. (1992). Predictors and consequences of childhood depressive symptoms: A 5-year longitudinal study. Journal of Abnormal Psychology, 101,405-422. Nurcombe, B., Seifer, R., Scioli, A., Tramontana, M. G., Grapentine, W. L., & Beauchesne, H. C. (1989). Is Major Depressive Disorder in adolescence a distinct entity? Journal of the American Academy of Child and Adolescent Psychiatry, 28, 333-342. OwascheI, H., Puig-Antich, J., Chambers, W., Tabrizi, M. A., &Johnson, R. (1982). Retrospective assessment of prepubertal major depression with the Kiddie-SADS-E. Journal of the Amen'can Academy of Child Psychiatry, 21, 392-397. Petersen, A. C., Cornpas, B. E., & Brooks-Gunn, J. (1992). Depression in adolescence: Current knowledge, research directions, and implicationsfor programs and policy. Washington DC: Carnegie Council on Adolescent Development. Petersen, A. C., Compas, B. E., Brooks-Gunn, J., Stemrnler, M., Ey, S., & Grant, K. E. (1993). Depression in adolescence. American Psychologist, 48, 155-168.

22 7

Petersen, A. C., Sarigiani, P. A,, & Kennedy, R. E. (1991). Adolescent depression: Why more girls? Journal of Youth and Adolescence, 20, 247-271. Petersen,A. C., Schulenberg, J. E.,Abramowitz, R. H., Offer, D., & Jarcho, H. D. (1984). A Self-Image Questionnaire for Young Adolescents (SIQYA): Reliability and validity studies. Journal of Youth and Adolescence, 13, 93-1 11. Phares, V., & Compas, B. E. (1992).The role of fathers in child and adolescent psychopathology: Make room for daddy. Psychological Bulletin, I1 1,387412. Piacentini, J., Shaffer, D., Fisher, P. W., SchwabStone, M., Davies, M., & Gioia, P. (1993). The Diagnostic Interview Schedule for ChildrenRevised Version (DISC-R): 111. Concurrent criterion validity. Journal of the American Academy of Child and Adolescent Psychiatry, 32, 658-665. Poznanski, E. O., & Mokros, H. B. (1994). Phenomenology and epidemiology of mood disorders in children and adolescents. In W. M. Reynolds & H. F. Johnston (Eds.), Handbook of depression in children and adolescents (pp. 19-40). New York: Plenum Press. Poznanski, E., Mokros, H., Grossman, J., & Freeman, L. (1985). Diagnostic criteria in childhood depression. American Journal of Psychiatry, 142, 11681173. Prieto, S. L., Cole, D. A., & Tageson, C. W. (1992). Depressive self-schemas in clinic and nonclinic children. Cognitive Therapy and Research, 16, 521-534. Puig-Antich, J., & Chambers, W. (1978). The Schedule for Affective Disorders and Schizophrenia for School-Age Children (Kiddie-SADS).New York: New York State Psychiatric Institute. Puig-Antich, J., Chambers, W., Halpem, F., Hanlon, C., & Sacher, E. J. (1979). Cortisol hypersecretion in prepubertal depressive illness: A preliminary study. Psychoneuroendocrinology, 4, 191-197. Puig-Antich, J., Perel, J. M., Lupatkin, W., Chambers, W. J., Shea, C.,Tabrizi, M. D., & Stiller, B. (1979). Plasma levels of imipramine (IMI) and desmethylimipramine (DMI) and clinical response in prepubertal major depressive disorder: A preliminary report. Journal of American Academy of Child Psychiatry, 18, 616627. Quay, H. C., & Peterson, L C . (1983). Interim manual for the Revised Behavior Problem Checklist. Unpublished manuscript, University of Miami. Radloff, L. S. (1977). The CES-D Scale: A self-report depression scale for research in the general population. Applied Psychological Measurement, I , 385-401. Rey, J. M., & Morris-Yates, A. (1991). Adolescent depression and the Child Behavior Checklist. Journal of the American Academy of Child and Adolescent Psychiatry, 30, 423-427. Rey, J. M., & Morris-Yates,A. (1992). Diagnostic accuracy in adolescents of several depression rat-

228

P A R T 111. E M O T I O N A L A N D S O C I A L D I S O R D E R S verszon 2.3 (DISC 2.3). Unpubl~shed manuscript, New York State Psychiatric Institute, New York P Shaffer, D , Schwab-Stone, M , F~sher, W , Cohen, P , P~acentlni, Davies, M , Conners, C K., & J., Regier, D (1993).'The Diagnostic Interview Schedule for Children-Revised Verslon (DISC-R) I. Preparation, field testlng, Interrater rellablllty, and acceptab~lity Iounzal of the American Academy of Child and Adolescent Psychzatv, 32,643-650. Shaffer, D , Schwab-Stone, M., Fisher, P., Davies, M , Piacentini, J., & G ~ o i aP. (1988). Results of , a field trlal and proposals for a new instrument (DISC-R) (Grant Nos. M H 36971 & M H C R C 30906-10) Bethesda, MD: National Institute of Mental Health. Shelton, R. C., Hollon, S. D., Purdon, S. E., &Loosen, P T. (1991) Biological and psychological aspects of depression. Behavior Therapy, 22, 201-228. Skinner, E. (1995). Motivation, coping and control. Newbury, CA. Sage. Smucker, M. R ,Cralghead, W. E , Cralghead, L W., & Green, B. J (1986). Normative and rellabllity data for the Children's Depression Inventory lournal ofAbnoma1 Child Psychology, 14,25-39. Spitzer, R. L., Endicott, J., & Robins, E (1978a). Research Diagnostic Criteria: Rationale and reliabil~ty. Archives of General Psychiatry, 35, 773-782. Sp~tzer, L , Endicott, J., & Robins, E. (1978b). ReR. search Diagnostic Cntena for a selected group of functional disorders (3rd ed.). New York: New York State Psych~atricInstitute. Tisher, M., & Lang. M. (1983). The Children's Depression Scale: Review and further developments. In D. P. Cantwell & G . A. Carlson (Eds.), Afiective disorders in ch~ldhood and adolescence: An update (pp. 181-202). New York: Spectrum. Watson, D., &Clark, L. A. (1984). Negative affectivity: to The d~sposition experience aversive emotional states. Psychological Bulletin, 96, 465-490. Watson, D., & Clark, L. A. (1992). Affects separable and inseparable: O n the hierarchical arrangement of the negative affects. journal of Personality and Social Psychology, 62, 489-505. Watson, D., & Kendall, P C . (1989). Common and dlfferentlatingfeatures of anxiety and depression: Current findlngs and future directions In P. C Kendall & D Watson (Eds.),Anxietyand depression: Distinctive and overlapping features (pp. 493-508). New I'ork: Academlc Press Watson, D., & Tellegen, A. (1985) Toward a consensual structure of mood. PsychologicalBullet~n, 98, 219-235. Wenstein, S. R., Noam, G . G , Grimes, K , Stone, K., & Schwab-Stone, M (1990). Convergence of DSM-I11 diagnoses and self-reported symptoms in child and adolescent inpatients. journal of the American Academy of Child and Adolescent Psychiatry, 29,627-634. Weiss, B , Weisz, J. R., Politano, M., Carey, M., Nelson, W . M., & F ~ n c hA. J. (1991). Develop, mental differences ~n the factor structure of the

ing scales extracted from a general purpose behavlor checkl~st.Ioumal of Affective D~sorders, 26, 7-16. Reynolds, W. M (1986). Reynolds Adolescent Depresslon Scale Odessa, FL: Psychological Assessment Resources. Reynolds, W M. (1989). Reynolds Child Depression Scale. Odessa, FL: Psychological Assessment Resources. Reynolds, W. M. (1994) Assessment of depression in children and adolescents by self-report questionnaires. In W . M. Reynolds & H F. Johnston (Eds ), Handbook of depress~onin children and adolescents (pp. 209-234). New York. Plenum Press. Roberts, R. E., Lewinsohn, P. M., & Seeley, J. R. (1991). Screening for adolescent depression: A comparison of depression scales. Ioumal of the Amencan Academy of Child and Adolescent Psychiatry, 30, 58-66. Robinson, N. S., Garber, J., & Hilsman, R (1995). Cognitions and stress: Direct and moderating effects on depressive versus externalizing symptoms during the junior high school transition. Ioumal of Abnormal Psychology, 104, 453-463. Rohde, p., Lewinsohn, P. M., & Seeley, J. R. (1991). Comorbidity of unipolar depression: I1 Comorbidity with other mental disorders in adolescents and adults. Ioumal of Abnormal Psychology, 54, 653-660. Rudolph, K. D., Hammen, C , & Burge, D. (1994). Interpersonal functioning and depressive symptoms in childhood Addressing the issues of specificity and comorbidity. Ioumal ofAbnormal Child Psychology, 22, 3 55-371. Saylor, C F., Finch, A. J., Spirito, A,, & Bennett, B. (1984). The Children's Depression Inventory: A systematic evaluation of psychometric properties journal of Consulting and Clznical Psychology, 52, 955-967. Schwab-Stone, M., Flsher, P. W., Piacentini, J , Shaffer, D., Davies, M., &Briggs, M. (1993). The Diagnostic Interview Schedule for ChildrenRevised Vers~on (DISC-R): 11. Test-retest rellability. Ioumal of the American Academy of Child and Adolescent Psychiatry, 32, 651-657. Selfer, R., Nurcombe, B., Scioli, A,, & Grapentine, W. L. (1989). Is Major Depressive Disorder in childhood a dlstinct diagnostic entity? \oumal of the American Academy of Child and Adolescent Psychiatry, 28, 93 5-941. Sellgman, M. E. P., Peterson, C., Kaslow, N J., Tanenbaum, R. L.,Nloy, L. B., &Abramson,L.Y. (1984). Attribut~onal styles and depressive symptoms among children. Joumal ofAbnormal Psychology, 93, 235-238. Shaffer, D., Fisher, P., Dulcan, D., Davies, M., Piacentlni, J., Schwab-Stone, M., Lahey, B., Bordon, K., Jensen, P., Bud, H., Canino, B., & Rejier, D. (1992) National Institute of Mental Health Diagnostic Interview Schedule for Children

Chapter 4. Depression in Children and Abksrents Children's Depression Inventory. Psychological Assessment: A Joumal of Consulting and Clinical Ps)lchology, 3, 38-45. Weissman, M . M., Gammon, G. D., John, K., Merikangas, K. R., Warner, V., Prusoff, B. A,, & Sholornskas, D. (1987). Children of depressed parents: Increased psychopathology and early onset of major depression. Archives of General Psychiatry, 44, 817-853. Weisz, J. R. (1986). Understanding the developing understanding of control. In M. Perlmutter (Ed.), Minnesota Symposium on Child Psychology: Vol. 18. Social cognition (pp. 2 19-278). Hillsdale, NJ: Erlbaum. Weisz, J. R., Sweeney, L., Proffitt, V., & Carr, T. (1993). Control-related beliefs and self-reported depressive symptoms in late childhood. Journal of Abnormal Psychology, 102,411-418. Weisz, J. R., Weiss, B., Wasserman, A. A,, & Rintoul, B. (1987). Control-related beliefs and depression among clinic-referred children and adolescents. journal of Abnormal Psychology, 96, 149-1 58.

229

Whitman, P. B., & Leitenberg, H. (1990). Negatively biased recall in children with self-reported symptoms of depression. Journal offinorma1 Child Psychology, 18, 15-27. Williams, C. L., & Butcher, J. N. (1989a). An MMPI study of adolescents: I. Empirical validity of the standard scales. Psychological Assessment: A Journal of Consulting and Clinical Psychology, I , 2 51259. Williams, C . L., & Butcher, J. N. (1989b). An MMPI study of adolescents: 11. Verification and limitations of code type classifications. Psychological Assessment: A Joumal of Consulting and Clinical Psychology, 1, 260-265. Wolfe, V. V., Finch, A. J., Saylor, C . F., Blount, R. L., Pallmeyer, T. P., & Carek, D. J. (1987). Negative affectivity in children: A multitrait-multimethod investigation. Journal of Consulting and Clinical Psychology, 55(2), 245-250. World Health Organization. (1992). lnternationaf classification of diseases and health-related problems (10th revision). Geneva: Author.